Do Not Delete
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
by
Laura A. Heymann∗
How do people talk when they talk about trademarks? If trademarks have
become, as linguist Geoffrey Nunberg suggests, our “new global tongue,”
perhaps we should pay greater attention to the grammar we use when we
talk about them. We use “Coke” to refer to the Coca-Cola beverage in the
North, and “coke” to refer to any kind of soda in the South, yet we still
manage to get the drinks we desire. We use trademarks as verbs—we
“xerox” a document or “tivo” a television program—without losing sight
of the fact that “Xerox” and “TiVo” are brands of particular products.
We use trademarks as metaphor and as slang—“Kleenex,” for example,
has been used as street language for ecstasy—without changing our
opinion of the products to which they relate. And yet, like overly academic
grammarians, courts and trademark owners often rely on linguistic
structures and rules in trademark law, telling consumers how to use and
pronounce the names of products and services with which they engage
and defining rights based on outmoded assumptions about conversations
around brands. Thus, as with the debate over the proper role of
dictionaries, trademark law might benefit from a more direct
consideration of its role in creating language—in other words, whether it
should be prescriptive (and define proper word usage) or descriptive (and
reflect common word usage). Incorporating linguistic theory on language
formation helps to begin this inquiry.
I.
II.
III.
INTRODUCTION ....................................................................... 1314
HOW LANGUAGE DEVELOPS: PRESCRIPTIVIST VERSUS
DESCRIPTIVIST APPROACHES................................................ 1322
THE LANGUAGE OF TRADEMARKS ....................................... 1329
A. Rule #1: Trademarks May Not Be Used Metaphorically ............. 1330
B. Rule #2: Trademarks Should Not Appear in Lowercase .............. 1337
© Laura A. Heymann 2010.
∗
Associate Professor of Law, College of William & Mary—Marshall-Wythe School
of Law. Many thanks to Mark Badger, Robert Brauneis, Trotter Hardy, Guy Heath,
Susan Mandiberg, Paul Marcus, Joseph Scott Miller, Carla Oakley, Stephen Scharf,
Jessica Silbey, Gerald Tschura, Rebecca Tushnet, and the participants at the “Who Is
the Reasonable Person?” symposium at Lewis & Clark Law School for helpful
comments and to Michael Adams and Anne Charity Hudley for useful suggestions
from a linguist’s perspective. Thanks also to Brad Bartels, Kristen Brown, Taylor
Davidson, and Will Versfelt for research assistance and to the staff of the Lewis &
Clark Law Review for their hard work.
1313
Do Not Delete
1314
IV.
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
C. Rule #3: A Trademark Must Always Be Used as an Adjective,
Never as a Noun (Including Possessive or Plural Forms)
or as a Verb ............................................................................. 1343
CONCLUSION ............................................................................ 1349
I.
INTRODUCTION
Tort law’s reasonable person is a measure of conduct—the activities
in which one engages, the choices one makes and, in trademark law, the
way in which one processes information. In negligence law, the
reasonable person is the embodiment of the law’s requirements: One is
1
negligent if one has failed to act in accordance with his or her actions,
and one is deemed not negligent if one has acted as he or she would
have done. In some cases, the jury may be asked simply to assess a party’s
actions in light of its own experience, taking into account some
characteristics (such as physical infirmities) and not others (such as
2
intelligence). In other cases, a court will take an economic approach by
3
assuming that the reasonable person takes cost-effective precautions. In
still other cases, a party’s actions will be assessed in light of external
4
standards, such as industry practice or a relevant statute. In all of these
instances, however, negligence law’s reasonable person standard is
largely prescriptive. Except in special circumstances, the law does not
base liability on whether a party acted according to his own preferences
or limitations, nor does the law expect litigants in most negligence cases
to present surveys of individuals asking how they would have acted under
similar circumstances. Rather, the law constructs an ideal reasonable
person as a standard by which to gauge the propriety of a party’s
conduct. A party is thus expected to comply with this standard or to train
herself to do so; in other words, the standard establishes a normatively
5
“correct” way of acting. To be sure, the standard incorporates some
1
Characterizing the reasonable person as either male or female may well have
ramifications with respect to whether the party’s activity is deemed to be reasonable.
See generally Margo Schlanger, Gender Matters: Teaching a Reasonable Woman Standard in
Personal Injury Law, 45 ST. LOUIS U. L.J. 769 (2001).
2
RESTATEMENT (SECOND) OF TORTS § 283C cmt. a (1965) (“So far as physical
characteristics are concerned, the hypothetical reasonable man may be said to be
identical with the actor.”); id. § 289 cmt. n (noting that the reasonable person
standard does not incorporate the defendant’s “attention, perception, memory,
knowledge of other pertinent matters, intelligence, and judgment”).
3
See United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947)
(Hand, J.) (using cost-benefit analysis of negligence).
4
See The T.J. Hooper, 60 F.2d 737, 740 (2d Cir. 1932) (rejecting argument that
industry standard should control); Tedla v. Ellman, 19 N.E.2d 987, 991–92 (N.Y.
1939) (holding that although duty is generally established by applicable statute,
plaintiffs were justified in noncompliance).
5
As Mayo Moran has noted, the risk of personification is that the reasonable
person’s assumed characteristics “tend to mirror those of the privileged in our
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1315
descriptive elements. To the extent that the law’s determination of what
constitutes reasonable care acknowledges evidence from the field—
jurors’ own experiences, industry standards, and the like—the standard
will reflect how individuals actually act (or, at least, how they believe they
would act). Moreover, in certain instances, such as medical malpractice
cases, where the industry standard is dispositive, industry participants
6
have the ability to shape the standard more directly. But the fact that the
reasonable person is an idealized standard that determines liability
means that it cannot escape its prescriptivist nature. It purports to reflect
what the ordinary person would do but, by incorporating a sense of
7
judgment, ultimately dictates what actors should do.
In trademark law, the reasonable person plays a different role.
Because the standard of fault in a trademark infringement or dilution
case is strict liability, the propriety of a defendant’s actions does not turn
on whether the defendant acted reasonably, although certain defenses
8
may incorporate the related concepts of fairness and good faith. Instead,
trademark law’s reasonable person is the person whom trademark law
9
purportedly benefits: the “reasonably prudent consumer.” The
reasonably prudent consumer provides the context against which the
defendant’s actions are judged, rather than the standard of liability itself,
and so at least purports to be descriptive in nature. In infringement
cases, for example, courts ask whether the defendant’s use of the
plaintiff’s mark would be likely to cause a reasonably prudent consumer
to be confused into thinking that the defendant’s goods or services are
produced by, authorized by, or sponsored by the plaintiff trademark
10
holder, rendering acceptable those uses that confuse only unreasonable
society.” MAYO MORAN, RETHINKING THE REASONABLE PERSON: AN EGALITARIAN
RECONSTRUCTION OF THE OBJECTIVE STANDARD 306 (2003).
6
See James Gibson, Doctrinal Feedback and (Un)Reasonable Care, 94 VA. L. REV. 1641,
1654–58 (2008).
7
Cf. MORAN, supra note 5, at 307 (discussing the “endemic confusion between
the normative and descriptive components of the [reasonable person] standard”). Of
course, the fact that the reasonable person is a standard means that its contours can
change over time. PETER M. TIERSMA, LEGAL LANGUAGE 79 (1999).
8
See, e.g., 15 U.S.C. § 1115(b)(4) (2006) (providing a defense to infringement if
the defendant’s use of the mark at issue is a use “which is descriptive of and used
fairly and in good faith only to describe the goods or services of such party, or their
geographic origin”). The standard of fault is not strict in a secondary liability case.
See, e.g., Inwood Labs., Inc. v. Ives Labs., Inc., 456 U.S. 844, 854 (1982).
9
This individual is also referred to in the case law as the “reasonable consumer,”
the “reasonable purchaser,” the “reasonably prudent purchaser,” or the “reasonably
careful purchaser,” among other similar terms. See, e.g., Int’l Ass’n of Machinists &
Aerospace Workers v. Winship Green Nursing Ctr., 103 F.3d 196, 201 (1st Cir. 1996)
(using “reasonably prudent purchasers exercising ordinary care”); Upjohn Co. v.
Schwartz, 246 F.2d 254, 262 (2d Cir. 1957) (using “reasonably careful purchasers”);
Royal Typewriter Co. v. Cachelin, 127 F. Supp. 173, 175 (S.D.N.Y. 1954) (using both
“ordinarily prudent purchasers” and “ordinary careful purchaser”).
10
See, e.g., Dreamwerks Prod. Grp. v. SKG Studio, 142 F.3d 1127, 1129 (9th Cir.
1998) (“The test for likelihood of confusion is whether a ‘reasonably prudent
Do Not Delete
12/15/2010 10:02 PM
1316
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
11
consumers. In attempting to answer the confusion question, courts
typically, as in negligence cases, take some characteristics of actual
consumers into account: whether they are sophisticated or
unsophisticated; whether they are under pressure to select goods quickly
or proceed more leisurely; whether they are capable of thoughtfully
12
making decisions in the marketplace or tend to act heedlessly. But,
unlike in negligence cases, courts in trademark cases will sometimes
consider evidence of the reasonable consumer’s cognitive process, such
as surveys, expert testimony, or third-party uses such as media reports or
dictionary entries. In this way, the reasonable person in trademark law
aims, at least, to be more descriptive than prescriptive: It attempts to
reflect how consumers actually think about the marks in question,
without conveying a normative judgment as to whether consumers should
13
think about the marks in such a manner. This is not to say that judges
consumer’ in the marketplace is likely to be confused as to the origin of the good or
service bearing one of the [plaintiff’s] marks.”).
11
Courts may also reference the reasonable consumer in dilution cases, in which
the question is whether the defendant’s use of the plaintiff’s mark is likely to dilute
the plaintiff’s mark through blurring or tarnishment, but they appear to be just as
likely to rely on the semantic association between the two marks. See, e.g., V Secret
Catalogue, Inc. v. Moseley, 605 F.3d 382, 385, 388 (6th Cir. 2010) (holding that
federal anti-dilution law creates a presumption “that a new mark used to sell sex
related products is likely to tarnish a famous mark if there is a clear semantic
association between the two,” but noting that the presumption may be rebutted by
evidence “in the form of expert testimony or surveys or polls or customer testimony”);
Visa Int’l Serv. Ass’n v. JSL Corp., 610 F.3d 1088, 1091 (9th Cir. 2010) (“[A] plaintiff
seeking to establish a likelihood of dilution is not required to go to the expense of
producing expert testimony or market surveys; it may rely entirely on the
characteristics of the marks at issue.”). But see Eli Lilly & Co. v. Natural Answers, Inc.,
86 F. Supp. 2d 834, 852 (S.D. Ind. 2000) (“For blurring to occur, there must be some
mental association in the reasonable consumer’s mind between the plaintiff’s and the
defendant’s uses of the mark.”).
12
Thus, among the various factors considered by courts in determining whether
the defendant’s actions are likely to cause confusion among consumers, the question
of the “reasonably prudent consumer” is likely to arise most directly in considering
consumer sophistication. See, e.g., Brookfield Commc’n, Inc. v. W. Coast Entm’t
Corp., 174 F.3d 1036, 1060 (9th Cir. 1999) (noting that the reasonably prudent
consumer is expected to be “more discerning—and less easily confused—when he is
purchasing expensive items”). For an overview of courts’ and commentators’ views on
consumer sophistication in this regard, see Thomas R. Lee, Glenn L. Christensen &
Eric D. DeRosia, Trademarks, Consumer Psychology, and the Sophisticated Consumer, 57
EMORY L.J. 575, 577–78 (2008).
13
See, e.g., E. I. DuPont de Nemours & Co. v. Yoshida Int’l, Inc., 393 F. Supp. 502,
510 (E.D.N.Y. 1975) (“In weighing the evidence of likelihood of confusion, the court
must strive to place itself in the shoes of a prospective purchaser. In this role, the
court does not act as an enlightened educator of the public but takes into account
the mythical ordinary prospective purchaser’s capacity to discriminate as well as his
propensity for carelessness.”); Barton Beebe, An Empirical Study of the Multifactor Tests
for Trademark Infringement, 94 CAL. L. REV. 1581, 1645 (2006) (suggesting that the
multifactor test of the likelihood of consumer confusion “is essentially a substitute for
empirical work”). The goal of descriptiveness may be more aspirational than actual.
See id. at 1641–42 (stating that although survey evidence is widely believed to be
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1317
are immune from the tendency to view issues from their own
14
perspectives, or that trademark law should not adopt a more
15
prescriptivist approach by conveying expectations to consumers. Nor is
it to ignore the fact that consumers are not a monolithic entity, and so
there will always be some consumers who respond differently from
others. But, on the whole, trademark cases purport to reflect courts’
desire to respond to conditions in the marketplace rather than to create
them ex ante.
There is one aspect, however, in which trademark law’s view of the
reasonable person embodies a tension between prescription and
description: in determining whether a word can function as a trademark
16
at all. On the one hand, the ultimate answer to this question is whether
a word will have or continues to have trademark significance to
consumers in the relevant market, a question that is typically answered by
reference to empirical evidence. On the other hand, this inquiry is in
service of a more prescriptive goal of trademark holders: to shape the
highly persuasive in trademark infringement cases, the case law does not accord with
this view); Graeme B. Dinwoodie, What Linguistics Can Do for Trademark Law, in TRADE
MARKS AND BRANDS: AN INTERDISCIPLINARY CRITIQUE 140, 157 (Lionel Bently et al. eds.,
2008) (noting that “linguistic understanding of key terms of art in trademark law
illustrates the inevitably prescriptive content of supposedly descriptive assessments of
trademark claims”).
14
See, e.g., Londontown Mfg. Co. v. Cable Raincoat Co., 371 F. Supp. 1114, 1118
(S.D.N.Y. 1974) (“In the trademark field, the precedents are so numerous and varied
that resort to categories is often difficult and unrewarding. That an infringer is the
Devil quoting Scripture sometimes cannot be determined until it is found that he is
an infringer. In this field of law more than almost any other, the reaction of the trial
judge as a person often concludes the issue. And sometimes an analogue can be
found for a decision either way, leaving the Court with the rather comfortable feeling
that it is, after all, supported by precedent.”). See also FRANK SCHECHTER, THE
HISTORICAL FOUNDATIONS OF THE LAW RELATING TO TRADE-MARKS 166 (1925) (noting
that “the so-called ordinary purchaser changes his mental qualities with every judge”)
(internal quotation marks omitted); Laura A. Heymann, The Reasonable Person in
Trademark Law, 52 ST. LOUIS U. L.J. 781, 783 (2008).
15
See, e.g., KP Permanent Make-Up, Inc. v. Lasting Impression I, Inc., 543 U.S.
111, 121 (2004) (noting that “some possibility of consumer confusion must be
compatible with fair use” of a mark); Stacey L. Dogan & Mark A. Lemley, The
Merchandising Right: Fragile Theory or Fait Accompli?, 54 EMORY L.J. 461, 488 (2005)
(contending that, in some trademark cases, “the law should act as a norms creator,
setting aspirational goals rather than responding to current consumer expectations”);
Alan Durant, “How Can I Tell the Trade Mark on a Piece of Gingerbread from All the Other
Marks on It?”: Naming and Meaning in Verbal Trade Mark Signs, in TRADE MARKS AND
BRANDS, supra note 13, at 107, 120 (considering linguistics’ relevance to the question
of “how far trade mark law should reflect public thinking about commercial
communication and how far it should be normative in relation to use of signs (as part
of commercial regulation and governance)”). As Joe Miller has pointed out to me,
courts in parody cases are also somewhat prescriptive in that they are essentially
telling consumers that they should “get the joke,” even if not all of them would
otherwise do so.
16
Nonlexical elements and trade dress can serve as and are also subject to many
of the claims raised herein. I focus in this Article only on word marks.
Do Not Delete
1318
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
language usage not only of competitors in the marketplace but of the
17
public more generally. For example, consumers, writers, and dictionary
editors are often reminded, more or less forcefully, that trademarks
should be used only as capitalized modifiers of common nouns, as in a
18
“Xerox photocopier.”
Trademark holders who communicate this view typically do so from
a position of perceived ownership and monopoly, one that is occasionally
19
reinforced by commentators. Although one would ordinarily not think
of language as something that can be owned, trademark doctrine
encourages this assumption, from the inescapable terminology of
property (which talks of “trademark owners” or “trademark holders”), to
the view that an invented word (such as “Xerox”) is a stronger trademark
than an existing word used in a trademark sense and thus deserves broad
legal protection, to the belief that appropriate enforcement activities
include not only attempts to stop competitors from engaging in uses of
marks that frustrate consumers’ purchasing decisions but also attempts to
limit the ability of consumers, writers, and others from using marks as
part of discourse. To extend the property metaphor further, then,
trademark law encourages trademark holders to think of enforcement as
akin to thwarting adverse possession: The trademark’s inviolability must
be defended against any and all encroachments, lest the right to own the
mark be relinquished to those who are using it for some personal benefit.
But while some uses of a mark are indeed infringing, not every use of
a mark is a legally relevant encroachment. To say that all uses of a
trademark inconsistent with its prescribed form will cause that trademark
to lose its original associations reflects a narrow view of how language
functions in society. That view, typically denominated by the linguistic
community as “prescriptive,” contends that communication benefits from
rules and directives supplied by authorities who educate speakers as to
20
correct uses of language. Descriptivists, by contrast, who occupy much
17
ROGER SHUY, LINGUISTIC BATTLES IN TRADEMARK DISPUTES 3 (2002) (discussing
trademark law’s attempt “to create a purism that tries to eliminate what it considers to
be the deviant varieties of language, a goal that is consistent with law’s innate need to
be prescriptive”). Shuy is a prominent expert witness in trademark cases.
18
Michael Atkins, Are Anti-Genericism Ads Effective?, SEATTLE TRADEMARK LAWYER
(Oct. 25, 2007, 9:32 PM), http://seattletrademarklawyer.com/blog/2007/10/26/areanti-genericism-ads-effective.html (discussing Xerox advertisement).
19
See, e.g., SHUY, supra note 17, at 2 (stating that “trademark law is about the right
to monopolize the use of language”); Stephen L. Carter, The Trouble with Trademark,
99 YALE L.J. 759, 769 (1990) (“Trademark protection matters because it allows a firm
to remove a word from the market language, in the sense that it allows the firm to
prevent others from using the word or anything confusingly similar to it. This creates
what might be called language exclusivity.”).
20
There is a certain resemblance between the standard economic justification
for many principles of liability in trademark law (reducing “search costs” or
“imagination costs” for consumers by eliminating conflicting signals) and the
justification offered by some prescriptivists for rules of language usage (to limit the
cognitive effort necessary to communicate). Compare, e.g., Qualitex Co. v. Jacobson
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1319
of the modern linguistics academy, take a more community-focused view
of language formation, in which language use is correct if it furthers
effective communication, without regard to its place in a language
hierarchy or compliance with certain rules of grammar. As a result,
context becomes vitally important, such that a lexical unit can have
multiple, and independent, meanings. Context may suggest a usage that
is formal or informal, standard or not standard, or traditional or modern,
but in all cases, language is, ultimately, the way in which reasonable
people communicate to promote understanding.
Trademark law has not yet fully resolved its place in this debate.
Although it purports in large part to be motivated by promoting
consumers’ understanding of commercial communications, it does not
strongly distinguish between uses of marks that frustrate understanding
and those that do not, often permitting trademark holders to control
21
expressive uses of a mark. Such distinctions would be more easily made
by focusing on the fact that a trademark is, quite simply, the proper name
22
of a brand of good or service. Trademark validity asks whether
consumers recognize a word as a proper name in context; trademark
infringement involves the confusing adoption of the same or similar
name by another for its goods or services; and trademark dilution
involves the use of a famous name in instances in which the law has
determined that only one entity is legally entitled to use that name for its
goods or services, regardless of the lack of confusion created by the
23
second use. A description of trademark law, then, that claims that the
Prods. Co., 514 U.S. 159, 163–64 (1995) (discussing search costs theory of trademark
law), and Ty Inc. v. Perryman, 306 F.3d 509, 511 (7th Cir. 2002) (discussing “higher
imagination cost[s]” as a possible justification for anti-dilution law), with David Foster
Wallace, Tense Present: Democracy, English, and the Wars over Usage, HARPER’S MAGAZINE,
Apr. 2001, at 39, 48–49 (noting that even though “[a] listener can usually figure out
what I really mean when I misuse infer for imply,” this usage may “require at least a
couple extra nanoseconds of cognitive effort, a kind of rapid sift-and-discard process,
before the recipient gets it”).
21
See, e.g., Mark A. Lemley & Mark McKenna, Irrelevant Confusion, 62 STAN. L.
REV. 413, 414 (2010).
22
Durant, supra note 15, at 114 (“What makes a sign ‘distinctive’ for the purpose
of trade mark law, from the language point of view, is essentially that it operates as a
name conventionally does: it attempts to make unique, unambiguous reference,
without elaboration or description, in a manner that will be successfully recognized in
a given context by the sign’s addressee (for trade mark law, the average consumer).”).
23
See Visa Int’l Serv. Ass’n v. JSL Corp., 610 F.3d 1088, 1091 (9th Cir. 2010)
(calling the association of two products with the same trademark the “quintessential
harm addressed by anti-dilution law”); 1 J. THOMAS MCCARTHY, MCCARTHY ON
TRADEMARKS AND UNFAIR COMPETITION § 3:6 (4th ed. Supp. 2008) (analogizing the
question “Who are you?” asked of a person, to which the usual response would be
one’s name, to the function performed by a trademark and distinguished from the
answer to the question “What are you?”). On the parallel between personal names
and trademark law, see WILLIAM M. LANDES & RICHARD A. POSNER, THE ECONOMIC
STRUCTURE OF INTELLECTUAL PROPERTY LAW 167 (2003); Laura A. Heymann, Naming,
Identity, and Trademark Law, 86 IND. L.J. (forthcoming 2010).
Do Not Delete
1320
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
doctrine is about establishing and preserving the meaning of marks rather
than certain associations with those marks risks unmooring the law from
its proper purpose. Names don’t have definitions as other words do;
“Marilyn Monroe” and “Microsoft,” for example, both have particular
references linked to biographical or historical data and evoke particular
associations, but one cannot provide a definition for either one in the
24
same way one can define the words “tree” or “table.” (In other words,
the names “Marilyn Monroe” and “Microsoft” each point to a single
reference in context, whereas “tree” and “table” can each be used to refer
to any one of a class of items.) Thus, whatever justifications can be
proffered for a prescriptivist view of language generally—such as
ensuring conformity to what are deemed “standard” (and often classbased) modes of speaking—those justifications are less compelling when
offered in the context of trademark law. If trademark law is really about
clarity in naming, reference, and associations, rather than about
meaning, it should discourage those uses that frustrate such references,
focusing instead on whether consumers view and use the mark in
question as the proper name of the plaintiff’s brand of product or
service. Attention to reference would avoid imbuing a more inventive
mark with property-like significance; rather the inquiry would be simply
whether, in context, consumers were likely to understand the word at
issue to have a naming function. In this way, trademark holders would
not be credited for challenging uses that do not reflect confusion as to
the mark’s referent. Uses of a mark in a verb form, for example (such as
“I xeroxed that document yesterday”), do not of themselves indicate that
the mark no longer has referential power, particularly given that
25
trademark holders engage in such uses themselves.
Ultimately, trademark law should, consistent with a descriptivist view
of language more generally, extend speakers’ fluidity with language into
the commercial sphere, reflecting consumers’ perceptions of language
24
JOHN I. SAEED, SEMANTICS 27 (2d ed. 2003) (“Names . . . are labels for people,
places, etc. and often seem to have little other meaning. It does not seem reasonable
to ask what the meaning of Karl Marx is, other than helping us to talk about an
individual.”). The treatment of trademarks in Webster’s Third was a matter of
considerable controversy. See HERBERT C. MORTON, THE STORY OF WEBSTER’S THIRD:
PHILIP GOVE’S CONTROVERSIAL DICTIONARY AND ITS CRITICS 215–23 (1994); Worth J.
Higgins, “Proper Names Exclusive of Biography and Geography”: Maintaining a
Lexicographic Tradition, 72 AM. SPEECH 381, 382 (1997). It is, admittedly, all too easy to
slip into the habit of describing trademarks as having meaning. See, e.g., Laura A.
Heymann, Metabranding and Intermediation: A Response to Professor Fleischer, 12 HARV.
NEGOT. L. REV. 201, 218–19 (2007).
25
Stuart Elliott, The Verb Treatment for an Investment House, N.Y. TIMES, Mar. 15,
2010, at B6 (reporting advertising campaign for the Vanguard Group that introduced
the verb “Vanguarding”); Geoffrey K. Pullum, Trademark Grammar, LANGUAGE LOG
(May 22, 2004, 11:42 AM), http://itre.cis.upenn.edu/~myl/languagelog/archives/
000943.html (suggesting that modern advertising encourages consumers to use
trademarks as language).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1321
26
rather than attempting to shape them. Drawing from linguistic theory
and modes of speaking in the nontrademark sphere, in which language
variations are a common characteristic of communication, can help
inform this process. The reasonable consumer, like the reasonable
person of tort law, must be at some level a judicial abstraction, a
27
necessary result of the time- and resource-limited process of litigation.
But courts need not disregard the fact that despite exhortations from
trademark holders, consumers use trademarks as language in a playful
sense, often without losing sight of the word’s reference to a brand of
good or service. The typical consumer does not always need to rely on
formal structures such as metaphorical hierarchies, parts of speech, or
rules of grammar to communicate effectively in every situation. Just as
reasonable speakers of English may easily accommodate both the use of
“hope” as a campaign slogan and “Hope” as a woman’s first name, the
reasonable consumer may well recognize that one can xerox something
on any photocopier but that only one company makes Xerox
28
photocopiers. The reasonable consumer likely understands that an Egg
McMuffin is a food item that can be ordered only at McDonald’s but that
a McJob can be held anywhere. Thus, trademark’s reasonable consumer
does not function prescriptively as a model of how other consumers
26
Cf. SHUY, supra note 17, at 195 (suggesting that linguistics is relevant to, inter
alia, questions relating to fanciful marks and the meaning understood by the
“ordinary person”).
27
1 KEITH ALLAN, LINGUISTIC MEANING 2 (1986) (“We ascribe to S[peaker] and
H[earer] the sort of capabilities and judgment attributed to the ‘reasonable man’ in
law; they are the hypothetical men-in-the-street who adopt whatever practice is
customary under the prevailing circumstances . . . .” (footnote omitted)); Jennifer
Davis, Locating the Average Consumer: His Judicial Origins, Intellectual Influences and
Current Role in European Trade Mark Law, 2 INTELLECTUAL PROP. Q. 183, 185 (2005)
(suggesting that “the existence of the average consumer [in European and U.K.
trademark cases] is itself a matter of trust rather than science”); id. at 198–99
(suggesting that greater equality in income and education in postwar Europe led
courts to assume a reasonable consumer); Dinwoodie, supra note 13, at 148 (noting
that the reasonable consumer must be “in large part a legal fiction that implements a
vision of the degree of consumer protection regulation that Congress and the courts
think appropriate without rendering commerce inefficient”).
28
Higgins, supra note 24, at 388 (“[A]lthough people often refer to any
photocopier as a xerox (the usual popular spelling—with a small x), still everyone
knows that there is also a copier whose brand name is Xerox.”). But see Ralph H.
Folsom & Larry L. Teply, Trademarked Generic Words, 89 YALE L.J. 1323, 1342 (1980)
(“Even when knowledgeable consumers understand a mark to be both a productcategory word and a trademarked brand name, their decisions respecting trade-offs
between quality and price may be hindered by their inability to determine whether
competitors’ products are indeed substitutable.”); Itamar Simonson, Trademark
Infringement From the Buyer Perspective: Conceptual Analysis and Measurement Implications,
13 J. PUB. POL’Y & MARKETING 181, 185 (1994) (discussing difficulties in distinguishing
consumer understanding of trademark status from consumer use of a trademarked
word). The difference in views may be explained by whether one is primarily a
linguist, a lawyer, or an economist and, relatedly, whether one is measuring
communicative effectiveness or purchasing decisions.
Do Not Delete
1322
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
should act; rather, trademark’s reasonable consumer helps us to
understand how trademark holders should act.
Reasonable speakers of English have an interest in effective
communication, and trademark law should assume that they talk about
trademarks with this goal in mind. An area of the law that purports to be
largely about the use of language should, as I discuss in Part II, take more
account of the way that language develops as it is spoken, rather than
encouraging trademark holders to impose rules of engagement. As I
discuss in Part III, such rules do not reflect the way that reasonable
people speak and, therefore, are unlikely to be obeyed except under
threat of legal sanction. Thus, it might be useful to abandon such efforts
in most circumstances, focusing plaintiffs’ and courts’ attention on the
relevant issue at hand: Does the available evidence indicate that
consumers understand the word in question to be a trademark, despite
29
uses that do not reflect a formalist approach to trademark grammar?
The answer to this question may not, in many cases, require additional
evidence; it may simply require a change in focus from a view of
trademarks as property to a view of trademarks as language. If, as one
commentator has suggested, the “reasonable person” represents, at its
30
core, “the module of shared commonsense world knowledge,” the law
should recognize that language is inherently a shared experience—no
less so when we are speaking the language of trademarks. In other words,
if trademark law purports to rely on how the reasonable consumer
engages with trademarks, it should be more willing to accommodate the
language the reasonable person actually uses.
II. HOW LANGUAGE DEVELOPS:
PRESCRIPTIVIST VERSUS DESCRIPTIVIST APPROACHES
The debate over how language develops generally involves two
31
opposing camps. The prescriptivist school takes an objective view of
language, asserting that there are correct meanings, pronunciations, and
grammars in a language, with rules that proper speakers of the language
are taught to follow. Language’s formalist nature, under this theory,
yields a set of “building blocks”: word parts that form words, which in
32
turn form sentences, which in turn form discourse. “Grammar” is then
“[t]he study of the building-block structure, the inherent properties of
29
Jerre B. Swann, The Validity of Dual Functioning Trademarks: Genericism Tested by
Consumer Understanding Rather Than by Consumer Use, 69 TRADEMARK REP. 357, 358
(1979) (contending that “the ultimate question is not the extent of consumer use in a
generic sense, but the extent of consumer understanding of a trademark sense”).
30
James Franklin, The Representation of Context: Ideas from Artificial Intelligence, 2
LAW, PROBABILITY & RISK 191, 198 (2003).
31
Although I focus in this Article on the English language, I do not mean to
suggest that different issues arise with respect to different languages (or, indeed, that
different issues do not arise).
32
GEORGE LAKOFF & MARK JOHNSON, METAPHORS WE LIVE BY 204 (1980).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1323
33
the parts, and the relationships among them.” Dictionaries and
grammar guides, according to this view, collect the appropriate rules and
34
are consulted as authoritative sources when debates arise.
The descriptivist school, by contrast, takes a subjective view of
language, rejecting the idea of a correct meaning, usage, or
pronunciation of a word. Descriptivists define language in terms of actual
use, such that meaning is derived from how people communicate rather
35
than from an external set of rules. This is not to say that this approach
dismisses structure or conventions; rather, it is simply that these
conventions are studied rather than imposed. Indeed, descriptivists
assume that both speakers and listeners are, to use the legal terminology,
“reasonable persons”: that they proceed according to the conventions of
the language in which they are conversing and that they share an interest
36
in effective and efficient communication. Because communication is
the goal of language usage, a particular usage cannot be said to be
incorrect so long as the audience for the communication can understand
37
what is being said. Descriptivist theorists thus view the appropriate goal
33
Id.; see also DAVID CRYSTAL, HOW LANGUAGE WORKS: HOW BABIES BABBLE, WORDS
CHANGE MEANING, AND LANGUAGES LIVE OR DIE 233 (2005) (describing a prescriptivist
grammar as “an account of the language’s possible sentence structures, organized
according to certain general principles”). Grammar might be then further broken
down into syntax (sentence formation) and morphology (word formation). See id. at
183.
34
William Strunk, Jr., and E.B. White’s The Elements of Style and the more recent
Eats, Shoots & Leaves by Lynne Truss have become popular volumes of the genre.
WILLIAM STRUNK JR. & E.B. WHITE, THE ELEMENTS OF STYLE (3d ed. 1979); LYNNE
TRUSS, EATS, SHOOTS & LEAVES: THE ZERO TOLERANCE APPROACH TO PUNCTUATION
(2004). See also JACK LYNCH, THE LEXICOGRAPHER’S DILEMMA 225 (2009) (describing
how Webster’s marketing department touted the dictionary as an authority because
“authority—even authoritarianism—sold dictionaries”).
35
LAKOFF & JOHNSON, supra note 32, at 217, 228. Thus, the objectivist meaning of
“grammar” is distinct from what some linguists refer to as a “generative grammar,”
the latter referring to the internal rules that speakers of English intuit to understand
why “the dog ran away” represents typical word order in English while “ran dog away
the” does not. See STEVEN PINKER, WORDS AND RULES: THE INGREDIENTS OF LANGUAGE 4
(1999).
36
ALLAN, supra note 27, at 3 (“There is (normally) a mutual expectation that
S[peaker] and H[earer] will behave reasonably towards one another, and will cooperate with one another, in their language interchanges as in other kinds of social
behaviour.”); PAUL GRICE, STUDIES IN THE WAY OF WORDS 26 (1989) (describing the
“Cooperative Principle” as expecting that a participant in a conversation will “[m]ake
[his or her] conversational contribution such as is required, at the stage at which it
occurs, by the accepted purpose or direction of the talk exchange in which [he or she
is] engaged”); Eve V. Clark & Herbert H. Clark, When Nouns Surface as Verbs, 55
LANGUAGE 767, 782–83 (1979) (noting that “indexical” expressions such as “that
woman” have meaning only in context, thus requiring a “moment-to-moment
cooperation” between speaker and listener).
37
The rules championed by prescriptivists are often characterized by
descriptivists as “standard English” or “formal English.” Geoffrey Nunberg, The Decline
of Grammar, ATLANTIC MONTHLY, Dec. 1983, at 31, 32 (“[T]he linguists have won over
a large part of the educational establishment, so that ‘correct English’ has come to
Do Not Delete
1324
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
of dictionaries and other reference tools as reflecting how the language is
38
spoken rather than setting down rules that must be followed. Although
dictionaries and similar reference works hold a certain place of authority
as sources of formal or standard English—witness the attention paid to
the words that are added to the leading dictionaries each year—the
ultimate goal of modern lexicographers is to act as “language
39
reporters,” not language legislators.
Of course, both schools derive meanings and grammar from some
external sources; the difference between the two is the nature of those
sources. Prescriptivists will often identify particular speakers of English as
authorities, relying on the works of well-regarded writers and other
40
educated users of language as examples of proper English. Descriptivists
rely more broadly on popular usage, using categories such as “standard”
and “nonstandard,” rather than “correct” and “incorrect,” to indicate
forms that have attained more widespread levels of usage. Thus, in the
descriptivist view, language is a fluid process, a living organism that
changes over time rather than a static, mathematical construct that
adheres to certain long-established rules; to a prescriptivist, rules provide
order and predictability, ensuring that speakers and listeners will share a
41
common understanding. For example, to a prescriptivist, the word
mean no more than ‘standard English,’ the English spoken by the educated middle
class.”).
38
See, e.g., SIDNEY I. LANDAU, DICTIONARIES: THE ART AND CRAFT OF LEXICOGRAPHY
240–72 (2d ed. 2001) (describing positions); LYNCH, supra note 34, at 113–15 (same);
Roy H. Copperud, English As It’s Used Belongs in Dictionary, EDITOR & PUBLISHER, Nov.
25, 1961, at 44, 44 (“Twenty minutes spent on the conclusions of any reputable
linguist in the last 25 years should convince even the most obtuse that the business of
a dictionary is to report how words are used, and not to prescribe or proscribe
meanings.”); Albert H. Marckwardt, Dictionaries and the English Language, 52 ENGLISH J.
336 (1963). The debate reached a fever pitch when the third edition of Webster’s New
International Dictionary was issued in 1961; many essays on the topic are collected in
DICTIONARIES AND THAT DICTIONARY (James Sledd & Wilma R. Ebbitt eds., 1962).
39
Strawberry Saroyan, In Land of Lexicons, Having the Last Word, N.Y. TIMES, Mar.
19, 2005, at B9 (quoting Erin McKean, editor-in-chief of the Oxford American
Dictionary). Courts and the Patent and Trademark Office will, on occasion, refer to
dictionaries when determining the trademark significance of a word. See, e.g., Am.
Heritage Life Ins. Co. v. Heritage Life Ins. Co. 494 F.2d 3, 11 n.5 (5th Cir. 1974)
(“The dictionary definition of the word is an appropriate and relevant indication of
the ordinary significance and meaning of words to the public.” (internal quotation
marks omitted) (using Webster’s Third New International Dictionary (1961))); MCCARTHY,
supra note 23, § 11:69 (discussing use of dictionaries by the Trademark Trial and
Appeal Board).
40
LYNCH, supra note 34, at 190 (“The job of the prescriptive grammarian is to
encourage writers to model their usage on the great writers of the past.”). Of course,
many of these writers did not consistently follow the rules for which they are cited as
authorities. See Geoffrey K. Pullum, 50 Years of Stupid Grammar Advice, THE CHRONICLE
REVIEW, Apr. 17, 2009, http://chronicle.com/article/50-Years-of-Stupid-Grammar/
25497.
41
Compare Nunberg, supra note 37, at 37 (“[W]e should no more ask linguistic
scientists to tell us what sounds best than we should ask economists to tell us which
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1325
“ain’t” would be characterized as not an English word, despite its facial
conformity with orthographical standards, while to a descriptivist, “ain’t”
is clearly part of the English language, given its long pedigree and usually
42
clear meaning, even if it might be characterized as nonstandard. The
element of shared understanding—of being able to characterize a word
as how reasonable people speak—is important even in a descriptivist
world. Teenagers who use “bad” to mean “good” might be hailed as
language innovators; someone who uses “bad” to mean “purple” will
almost certainly be deemed to be talking nonsense, at least until
43
sufficient numbers of people adopt a similar usage. This is not to say
that a modern linguist would advocate abandoning all grammar and
vocabulary education or that term papers will henceforth be written in
the same vocabulary as text messages. To the contrary, modern linguists
acknowledge that certain forms of communication are appropriate for
certain contexts—a business letter will be written differently from a
postcard, and a discussion between friends will sound different from a
discussion in a courtroom. But all this means is that speakers of English
use the language that is appropriate to the circumstances: more formal
English in more formal settings, and more casual English elsewhere, with
either one being “correct” only in particular circumstances.
It is true that users of dictionaries, having been told throughout their
schooling that certain language uses are incorrect, may not always grasp
this distinction. The continued vitality of works such as Eats, Shoots &
Leaves and similar grammar guides suggests that speakers of a language
use dictionaries and similar reference works not simply to learn the
meanings of words and structures with which they are unfamiliar but to
44
learn the correct meaning or pronunciation of a word. But even
distribution of property will be fairest; those matters are for us to decide.”), with Mark
Halpern, A War that Never Ends, ATLANTIC MONTHLY, Mar. 1997, at 19, 20 (responding
to Nunberg by asserting that arguments that linguistic rules are man-made “are no
more arguments against laws governing language usage than they are against laws
governing vehicular traffic”).
42
McCandlish Phillips, Webster Soups Up Its Big Dictionary, N.Y. TIMES, Sept. 7,
1961, at 37 (discussing inclusion of “ain’t” in Webster’s Third New International
Dictionary). Indeed, from a descriptivist view, a well-entrenched nonstandard usage
may, over time, become standard. LYNCH, supra note 34, at 15 (noting of certain
pronunciations that “[a]t some point the wrong version became right”).
43
PINKER, supra note 35, at 15–16. Certain race- and class-based judgments will
affect whether language variations are deemed “acceptable” and by whom; witness,
for example, the debate between those who characterize a certain language variation
as “African American Vernacular English” and those who characterize it as “Ebonics.”
See Geoffrey K. Pullum, Ideology, Power, and Linguistic Theory 3–4 (Dec. 13, 2006),
http://people.ucsc.edu/~pullum/MLA2004.pdf (discussing the debate).
44
Wilson Follett, Sabotage in Springfield: Webster’s Third Edition, THE ATLANTIC, Jan.
1962, at 73, 77 (“The fact that [a dictionary’s] compilers disclaim authority and
piously refrain from judgments is meaningless: the work itself, by virtue of its
inclusions and exclusions, its mere existence, is a whole universe of judgments,
received by millions as the Word from on high.”); Dwight MacDonald, Three Questions
for Structural Linguists, Or Webster 3 Revisited, in DICTIONARIES AND THAT DICTIONARY,
Do Not Delete
1326
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
characterizing a word usage as “correct” tells us nothing about how that
status is achieved. In modern lexicography, “correct” means “most
accepted usage”; it is not the equivalent of a pronouncement from
45
l’Academie française. We look to grammar guides to tell us how a
certain class of individuals speaks in certain situations; we would seem
foolish if we used these same manuals to guide our conversations in all
circumstances.
Admittedly, this formal/casual distinction does not map directly
onto trademark usage among consumers. Consumers likely do not have
an intuition that leads them to use the word “Xerox” differently in a
more formal setting from the way they would use it in a more casual one,
in the same way they might use “I am going to tell her” in the former and
“I’m gonna tell her” in the latter. But they do think about trademarks in
a related way in that they typically understand that context matters to
reference. If a reasonable user of English knows that the word “text” in
the sentence “Can you please text me your telephone number?” relates to
a different technology from the one denoted by the word “text” in the
sentence “All students must have the required text on the first day of
class,” that same speaker of English might well understand that the word
“Xerox” in the sentence “You should buy a Xerox rather than a Canon;
the Xerox is cheaper” has a different reference from the word “xerox” in
the sentence “Can you please xerox this document for me before you go
home?” It is when the first sentence becomes nonsensical that “Xerox”
fails to have any distinctive reference, not necessarily when the second
sentence becomes a matter of common usage.
Yet just as grammar guides aim to tell readers the proper ways to use
language, trademark owners expend effort on educating consumers,
writers, and others about rules of trademark usage that would mandate
the characterization of the second “xerox” sentence as incorrect. These
shibboleths arise from the belief that in order to prevent a slide to
46
generic status, a trademark must always be used in precisely the same
supra note 38, at 256, 257 (noting that Webster’s Third’s use of labels such as
“substandard,” “nonstandard,” and “slang” are in part prescriptive because “they
imply a value judgment”); Patrick E. Kilburn, Onward to Agincourt: Or, Once More Unto
the Breach, Dear Friends, in DICTIONARIES AND THAT DICTIONARY, supra note 38, at 265,
266 (refuting MacDonald by explaining that a dictionary’s label of a usage as
nonstandard is, in fact, descriptive).
45
Cf. TIERSMA, supra note 7, at 116 (“[W]hen Noah Webster approached Chief
Justice John Marshall for an endorsement of his dictionary, Justice Marshall is
reported to have refused, commenting that in America it was individuals rather than
public bodies that dictate language use.” (citing Shirley Brice Heath, A National
Language Academy?: Debate in the New Nation, 189 LINGUISTICS 9, 33 (1977))).
46
As to the appropriateness of the term “genericide,” see John Algeo, Five
Reference Books, 51 AM. SPEECH 144, 148 (1976) (reviewing THE NEW COLUMBIA
ENCYCLOPEDIA (William H. Harris & Judith S. Levey eds., 1975)) (calling “genericide”
“one of the worst constructed neologisms of recent times” because its word formation
suggests that it means “‘murder of genera (or, in this case, generic terms),’” rather
than its opposite).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1327
way, as “Xerox® photocopiers” and in no other form, because other uses
will suggest that the word no longer serves as a trademark. Trademark
holders presumably take these actions because courts deem the extent to
which a trademark holder has “policed one’s trademark” as an important
factor in determining trademark validity, including policing uses of the
47
mark beyond potentially confusing uses by competitors. This is both
curious and unfortunate: curious, because it seems to conflict with
assertions elsewhere by courts that exhortations in the form of
48
advertising cannot serve to transform a generic term into a trademark,
and unfortunate, because it encourages trademark holders to send ceaseand-desist letters to any entity using a trademark in an unorthodox
49
fashion, including media outlets and dictionary publishers, regardless of
47
See Ill. High Sch. Ass’n v. GTE Vantage Inc., 99 F.3d 244, 246 (7th Cir. 1996)
(Posner, J.) (“A serious trademark holder is assiduous in endeavoring to convince
dictionary editors, magazine and newspaper editors, journalists and columnists,
judges, and other lexicographically influential persons to avoid using his trademark
to denote anything other than the trademarked good or service.”); Cullman
Ventures, Inc. v. Columbian Art Works, Inc., 717 F. Supp. 96, 124 (S.D.N.Y. 1989) (“A
trademark owner’s efforts at policing its trademarks is further proof of the strength of
those marks.”); Charles R. Taylor & Michael G. Walsh, Legal Strategies for Protecting
Brands from Genericide: Recent Trends in Evidence Weighted in Court Cases, 21 J. PUB. POL’Y
& MARKETING 160, 164–65 (2002) (contending that the success rate for parties that
rely on “indirect evidence such as newspaper and trade journal articles and dictionary
definitions” is low but encouraging trademark owners to send letters to “media that
misuse the trademark”). But see Playboy Enters., Inc. v. Chuckleberry Pub., Inc., 486 F.
Supp. 414, 422–23 (S.D.N.Y. 1980) (“The owner of a mark is not required to police
every conceivably related use—thereby needlessly reducing non-competing
commercial activity and encouraging litigation—in order to protect a definable area
of primary importance.”). See also Deven R. Desai & Sandra L. Rierson, Confronting the
Genericism Conundrum, 28 CARDOZO L. REV. 1789, 1792 (2007) (contending that
trademark doctrine gives too much weight to “non-commercial and/or noncompetitive trademark use contexts” in determining the generic status of a mark).
48
See infra note 54.
49
Linguists generally maintain that dictionary editors should not be responsive
to such requests. See, e.g., LANDAU, supra note 38, at 407 (“The lexicographer has no
intention of depriving a trademark owner of the exclusive right to use a term; but he
is in the business of writing dictionaries and his overriding obligation is therefore to
dictionary users. He cannot allow any special-interest group to determine what goes
in his dictionary or how it is represented.”); Michael Adams, Lexical Property Rights:
Trademarks in American Dictionaries, VERBATIM, Winter 2005, at 1, 1 (urging dictionaries
to “record the flow of trademarks into the general American vocabulary”).
Commentators are divided as to the legal implications of dictionary action or
inaction. Compare, e.g., Deborah R. Gerhardt, Consumer Investment in Trademarks, 88
N.C. L. REV. 427, 494 (2010) (arguing in favor of a safe harbor from trademark
infringement for reference tools), with, e.g., Julius R. Lunsford, Jr., Trademarks and
Semantics: The Use and Misuse of Trademarks in Dictionaries and Trade Journals, 6 GA. L.
REV. 311, 334 (1972) (arguing in favor of a statutory remedy “designed to correct the
misuse and the incorrect definition of trademarks in dictionaries and other
publications”), and Gary C. Robb, Trademark Misuse in Dictionaries: Inadequacy of
Existing Legal Action and a Suggested Cure, 65 MARQ. L. REV. 179, 195 (1981) (arguing in
favor of a cause of action against any entity, including a dictionary publisher, “that
repeatedly uses a registered trademark in a manner which causes it to lose its
Do Not Delete
1328
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
whether such use serves to indicate that the mark has ceased to function
50
as the proper name of a brand of good or service. These letters, like
many other attempts to characterize the state of the law to the public, can
create both a legal and a linguistic feedback loop. From a legal
perspective, such letters risk expansion of the scope of trademark law. If
the letters’ recipients come to believe that the law permits only
authorized uses of trademarks, this belief may be reflected in surveys and
51
other assessments of consumer perceptions of trademark validity. From
a linguistic perspective, such letters may train consumers and the media
in how to talk about trademarks, even if this causes them to speak
unnaturally, which will then nevertheless feed back into courts’ views of
52
whether a term continues to serve as a trademark.
One cannot blame trademark owners in this regard: Their economic
interests compel them to resist language change of this sort, and courts
have, correspondingly, done little to discourage them from engaging in
broad enforcement efforts. Indeed, the nature of the judicial and
administrative process in trademark disputes, which results in a ruling or
opinion, conveys a prescriptivist’s sense of trademark use as something to
be legislated rather than reported. But while the ever-expanding
consumer marketplace may be in need of structure and rules in order to
significance as an indicator of origin”). (Lunsford was, at the time his article was
published, the assistant vice president in charge of trademarks and patents for the
Coca-Cola Company.) The European Union takes a different view of the duties of
dictionary publishers. See Council Regulation 40/94, art. 10, 1993 O.J. (EC)
(requiring revision of entries upon request of a trademark holder if an entry “gives
the impression that it constitutes the generic name of the goods or services for which
the trade mark is registered”). The International Trademark Association provides its
members with a form letter that can be sent to dictionary publishers. See Int’l
Trademark Ass’n, Misuse of Trademarks in Dictionaries and Style Guides,
http://www.inta.org/images/stories/downloads/InfoPub/dictionary and style guide
template letter.doc.
50
In a 2010 April Fool’s Day post, Google played on these conventions by
announcing that in light of the (humorous) decision of the city of Topeka, Kansas, to
change its name to “Google,” it would be changing its name to “Topeka”; the post
included a list of appropriate uses of the new name (“Before our blind date I did a
Topeka search on him”) and inappropriate uses (“Before our blind date I topeka’d
him”). Eric Schmidt, A Different Kind of Company Name, THE OFFICIAL GOOGLE BLOG
(April 1, 2010, 12:01 AM), http://googleblog.blogspot.com/2010/04/different-kindof-company-name.html.
51
William McGeveran, The Trademark Fair Use Reform Act, 91 B.U. L. REV.
(forthcoming 2010) (noting that threat letters “serve a broader signaling function, by
warning against all unauthorized uses of a mark”); Mark P. McKenna, Trademark Use
and the Problem of Source, 2009 U. ILL. L. REV. 773, 774 (2009) (“Consumer
expectations largely define trademark rights, yet those expectations are influenced by
consumers’ understanding (or misunderstanding) of the law.”).
52
Folsom & Teply, supra note 28, at 1353 (suggesting that media usage of marks
“may reflect not the public’s understanding of the trademarked word, but rather the
media’s own editorial policies—policies that inevitably reflect the existence of
numerous trademark-policing programs designed to pressure the media into ‘proper’
trademark usage”).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1329
function effectively, this need does not require the trademark law
equivalent of Strunk and White. If trademark law does indeed purport to
reflect how the reasonable consumer acts in the marketplace, that
reflection should incorporate the fact that language variation is an
inherent characteristic of the English language.
III. THE LANGUAGE OF TRADEMARKS
Trademark validity depends on how consumers interpret words.
Because trademarks are, as is typically described, indicators of source,
whether something functions as a trademark depends on whether
consumers interpret a word as functioning as the proper name of a
brand of product or service as opposed to communicating some other
information about the product. In the phrase “Hershey’s delicious
chocolate,” for example, the word “chocolate” communicates the type of
product at hand; the word “delicious” communicates a characteristic
about the product; and the word “Hershey’s” communicates the proper
name of the brand of chocolate (which identifies, explicitly or implicitly,
the company that makes or takes responsibility for the quality of the
candy). As with language generally, and particularly with names, much
depends on context. “Apple” on its own does not indicate trademark
status until one knows if the speaker is talking about fruit or computers,
just as “George Washington” may indicate the name of a president or the
name of a university, depending on the nature of the communication at
53
issue.
Courts and commentators have long recognized that trademark
meaning is a matter of consumer response. In other words, the mere fact
that a company has spent millions of dollars in advertising a brand or
sent innumerable cease-and-desist letters encouraging a particular use of
a claimed mark does not lead to trademark validity if those efforts fall on
54
deaf ears. The ultimate question, as Judge Hand noted many years ago,
“is merely one of fact: What do the buyers understand by the word for
55
whose use the parties are contending?” More specifically, as the
53
The graphical presentation of a word may also affect how consumers interpret
a term. See Thomas R. Lee et al., An Empirical and Consumer Psychology Analysis of
Trademark Distinctiveness, 41 ARIZ. ST. L.J. 1033, 1038 (2009) (contending that
graphical placement of the text is what communicates trademark status to consumers
and that even a generic word will be interpreted as a trademark given appropriate
graphical treatment).
54
See, e.g., Co-Rect Prods., Inc. v. Marvy! Adver. Photography, Inc., 780 F.2d 1324,
1332 (8th Cir. 1985) (“[D]esires or intentions of the creator . . . are irrelevant.
Instead, it is the attitude of the consumer that is important.”); Dupont Cellophane
Co. v. Waxed Prods. Co., 85 F.2d 75, 81 (2d Cir. 1936) (“It, therefore, makes no
difference what efforts or money the DuPont Company expended in order to
persuade the public that ‘cellophane’ means an article of DuPont manufacture. So
far as it did not succeed in actually converting the world to its gospel it can have no
relief.”).
55
Bayer Co. v. United Drug Co., 272 F. 505, 509 (S.D.N.Y. 1921).
Do Not Delete
1330
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
Supreme Court phrased it some years later, a putative trademark holder
must demonstrate that the “primary significance of the term in the minds
56
of the consuming public is not the product but the producer.” If
consumers understand the word to be the proper name of a brand of
product or service, the word serves as a trademark (that is, as a reference
to the company that stands behind the good); if consumers understand
the word to be something else (for example, the common name of a
product or a characteristic of the product), the word does not serve as a
trademark in that context.
Despite the narrow focus of this inquiry, some courts, trademark
holders, and commentators have focused not on assessing the
understanding of the reasonable consumer but rather on adhering to
various prescriptivist rules of trademark grammar, concluding that failure
to follow these rules means, ipso facto, that the claimed trademark is
invalid. The underlying theme of these rules is that a trademark must be
represented in only one lexical form (without a plural suffix and with a
capital letter), as only one part of speech (as an adjective, not as a noun
or verb), and with a tightly constrained scope of references (and so not
used in any metaphorical or colloquial sense). But just as the sentence “I
wondered who she was referring to” continues to have meaning in
English despite its lack of conformity with prescriptivist rules of grammar,
a trademark can still be viewed as a trademark despite uses by consumers
and writers that fail to conform with these purported rules of trademark
grammar. I discuss some of these rules, and how our understanding of
reasonable (and common) modes of conversation subvert them, below.
A. Rule #1: Trademarks May Not Be Used Metaphorically
Both the prescriptivist and descriptivist views of language
incorporate a theory of metaphor—the phenomenon of “understanding
57
and experiencing one kind of thing in terms of another,” such as the
trope we use when we talk about time “marching on.” Metaphors are
such a fundamental part of communication that we likely are not even
58
aware of their use in many instances. Indeed, the concept of metaphor
is fundamental to how most trademarks work. Except for words invented
to serve as trademarks—such as “Kodak” and “Xerox”—all trademarks,
being words in the English language, operate on a level other than a
literal one in that they require consumers to use a familiar word or
56
Kellogg Co. v. Nat’l Biscuit Co., 305 U.S. 111, 118 (1938).
LAKOFF & JOHNSON, supra note 32, at 5.
58
Lee et al., supra note 53, at 1050 (asserting that “there is ample evidence that
people derive the semantic meanings of complex stimuli such as metaphors so rapidly
and effortlessly that the cognitive process can be called automatic”). Those favoring
an objective theory of language are likely to view metaphor as simply an alternative
correct meaning for the word, while descriptivists will judge the metaphor by whether
it fulfills the needs of participants to a communication for effective discourse. See
LAKOFF & JOHNSON, supra note 32, at 211.
57
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1331
59
expression in a new and initially unfamiliar context. The trademark
“Brawny” for paper towels, for example, is intended not only to serve as
the name of the product but also to encourage consumers to think of the
product in terms of human strength. The trademark “Apple” for
consumer electronics, while perhaps not designed to create a
metaphorical association between the company and fruit, was presumably
chosen because of the word’s appealing associations; alternatives with a
negative association (such as “Spinach” for the youth market), one can
60
assume, were not under consideration. Even a mark that more directly
describes the characteristics of the good or service with which it is
associated operates using a form of metaphor, so long as it is recognized
as a name rather than as a description, in that it encourages consumers
to associate the term with a particular company and its products rather
than with all products of that class generally (hence the reference in the
61
doctrine to “secondary meaning”). In short, a trademark works because
59
See Modular Cinemas of Am., Inc. v. Mini Cinemas Corp., 348 F. Supp. 578,
586 (S.D.N.Y. 1972) (“Words must be used in a ‘trademark sense’ in order to
constitute a mark. This means they must function in a fanciful, imaginative manner
when perceived by the public.”); United Lace & Braid Mfg. Co. v. Barthels Mfg. Co.,
221 F. 456, 461 (E.D.N.Y. 1915) (“Every good trade-mark is suggestive; once seen or
heard, its association with the product is readily fixed in the mind.”). Cf. SHUY, supra
note 17, at 37 (“It is also quite likely that all marks are suggestive in some sense,
otherwise they wouldn’t work at all.”). Some courts recognize the metaphorical
nature of trademarks but only as to suggestive marks. See, e.g., BigStar Entm’t, Inc. v.
Next Big Star, Inc., 105 F. Supp. 2d 185, 196 (S.D.N.Y. 2000) (“When choosing what
to call the article, the creator of the suggestive name meaningfully fixes upon
associational terms that will identify the product figuratively and will appeal to the
consumer by allusion and metaphor.”).
While invented words have no literal meaning, they are, presumably, created with
an eye and ear toward their linguistic effect. As Steven Pinker notes, companies
incorporate phonesthesia when they invent brand names “using faux-Greek and
Latinate neologisms built out of word fragments that are supposed to connote certain
qualities without allowing people to put their finger on what they are.” STEVEN
PINKER, THE STUFF OF THOUGHT: LANGUAGE AS A WINDOW INTO HUMAN NATURE 304
(2007) (citing Acura, Verizon, Viagra, and Altria as examples); see also JOHN ALGEO,
ON DEFINING THE PROPER NAME 58 (1973) (“To take words like Vercingetorix or
Popocatepetl as better examples of proper names than Sue and Dartmouth because the
former are arbitrary, at least to most English speakers, whereas the latter have some
descriptive value, is to ignore the obvious fact that most names have some motivation.
Those that do not are the exception, not the paradigmatic case.”).
60
Cf. MARTY NEUMEIER, THE BRAND GAP: HOW TO BRIDGE THE DISTANCE BETWEEN
BUSINESS STRATEGY AND DESIGN 84 (2006) (suggesting that trademarks that “use AngloSaxon words, or the names of people, tend to be high-imagery names, producing
vivid mental pictures that aid recall” (citing Apple Computer and Betty Crocker as
examples)); Barton Beebe, The Semiotic Analysis of Trademark Law, 51 UCLA L. REV.
621, 671 (2004) (categorizing such marks as catachrestic).
61
Durant, supra note 15, at 130, 134 (construing “descriptive” trademarks as
consistent with “a sense closer to ‘general communicative use’, and contrast[ing] with
specifically ‘naming’ use”); Dinwoodie, supra note 13, at 143 (“On the whole . . . it is
probably true that terms that are descriptive fail immediately to act as source
identifiers in the marketplace.”).
Do Not Delete
1332
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
it causes consumers to think of a particular lexical unit as a proper name
in addition to whatever literal meaning the word holds. The lexical unit
“camel” engenders no metaphorical associations when it is used in
connection with the desert animal; the same word used in connection
62
with cigarettes operates on an additional “dimension of meaning.”
Thus, trademark validity should depend on a fairly binary
distinction: Does the word operate on a literal level (as the common
name of a good or service) or on a metaphorical level (as the proper
name of a brand of good or service)? Trademark doctrine’s Abercrombie
63
hierarchy, however, rewards the creativity inherent in a mark, and in so
doing, imports Lockean property notions into what should be a linguistic
64
analysis. In his opinion for the Second Circuit, Judge Friendly set forth
“some basic principles of trademark law,” beginning with what he noted
were the “four categories of terms” from the case law that reflected
increasing degrees of “eligibility to trademark status and the degree of
protection accorded,” ranging from no protection at all to the strongest
degree of protection: (1) generic (words such as “bread” for the baked
good), (2) descriptive (words such as “Fresh” as a mark for bread),
(3) suggestive (words such as “Farmer’s Choice” as a mark for bread),
65
and (4) arbitrary or fanciful (such as “Rainbow Bread” or “Zazz Bread”).
Locating a claimed mark along this spectrum has now become a typical
first step in any trademark opinion, much as a grammarian might begin
66
his or her analysis by diagramming a sentence. The purpose of the
exercise, as suggested by Judge Friendly, is to determine whether the
word at issue functions as a trademark, such that “inherently distinctive”
62
Durant, supra note 15, at 130 (“[T]he capability of the sign to have more than
one dimension of meaning—to operate on more than one level at once, like a poetic
image—is almost certainly what the trade mark owner wanted it to do.”).
63
Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 9 (2d Cir. 1976).
64
Cf. Ill. High Sch. Ass’n v. GTE Vantage Inc., 99 F.3d 244, 246 (7th Cir. 1996)
(“[A] trademark is not nearly so secure an entitlement as a property right. It is mainly
just a designation of source and dies when it ceases to designate, for whatever reason
other than the culpable conduct of the defendant.” (citations omitted)).
65
Abercrombie, 537 F.2d at 9. This was not the only categorization that could have
been drawn from the existing case law. See, e.g., OWEN DAVIES TUDOR, A SELECTION OF
LEADING CASES ON MERCANTILE AND MARITIME LAW 573 (2d ed. 1868) (distinguishing
fanciful names from four other kinds of trademarks: (1) those denoting the place of
manufacture, (2) those denoting the manufacturer, (3) symbols, and
(4) combinations of the other four types). For discussions of the linguistic properties
of trademark neologisms, see Louise Pound, Trade-Name Irradiations, 26 AM. SPEECH
166 (1951), and Louise Pound, Word-Coinage and Modern Trade-Names, 4 DIALECT
NOTES 29, 40–41 (1913) (noting the increase in invented trademarks in the late
nineteenth and early twentieth centuries and expressing surprise that consumers did
not view such neologisms with distrust).
66
See, e.g., Lee et al., supra note 53, at 1035 n.4 (stating that the Abercrombie
hierarchy “has been embraced repeatedly by the Supreme Court as establishing the
governing framework for trademark distinctiveness under federal law”); id. at 1037
(asserting that “[j]udges, practitioners, and scholars reflexively and unquestioningly
have embraced the Abercrombie formulation of trademark distinctiveness”).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1333
marks (fanciful, arbitrary, and suggestive marks) are entitled to
protection ab initio, while descriptive marks are protectable only upon
proof of “secondary meaning”—that is, when the relevant consumer
market views the word as a mark rather than as a description of a
characteristic of the good or service. The more imaginative the
trademark is determined to be, according to the Abercrombie hierarchy,
67
the more protection it receives.
But, as other commentators have suggested, once it has been
determined that a word functions as a mark—that it can and does
function as a proper name of a product or service—its degree of
68
metaphor is irrelevant to its validity. Fanciful, arbitrary, suggestive, and
descriptive marks with secondary meaning are all deemed valid marks,
and the “degree of protection accorded” would seem to be, as a matter of
formalism, the same for each. If the question at hand is truly what
consumers understand “by the word for whose use the parties are
69
contending,” the answer need not depend on whether the word is
invented, suggestive of the product’s characteristics, or descriptive of
them. Rather, the answer should depend on whether the reasonable
consumer would understand the word in question—in light of
70
presentation, use, and context—to be functioning as a proper name.
Generic words and mere descriptions do not function in any metaphoric
sense at all. They are lexical units that can be defined and thus can be
said to have meaning, in contrast to proper names, which can be defined
71
only in terms of their referential function. Indeed, the case law on
which the Abercrombie court relied suggests this framework. Earlier courts
had not viewed trademark validity along a spectrum of distinctiveness,
67
See, e.g., Dreamwerks Prod. Grp., Inc. v. SKG Studio, 142 F.3d 1127, 1130 (9th
Cir. 1998) (rejecting argument that “dream” as used in the trademark “Dreamwerks”
is suggestive, noting that “‘[d]ream’ is used in too many different ways to suggest any
particular meaning to the reasonable consumer”); BigStar Entm’t, Inc. v. Next Big
Star, Inc., 105 F. Supp. 2d 185, 199 (S.D.N.Y. 2000) (“In this Court’s view, no subtlety
or intimation, no mental detour or thought gymnastics are demanded to span by
creative leap the small space evident here between the product and name and its
source.”). Note, however, that a court’s determination of whether a claimed mark is,
for example, suggestive may well depend on the vocabulary the court brings to the
exercise, knowledge that may or may not match that of the relevant consumer.
68
Beebe, supra note 60, at 671 (noting that eligibility for trademark protection
depends on whether the mark is distinctive of source or not); Lee et al., supra note
53, at 1038 (advocating elimination of Abercrombie hierarchy in favor of test that
renders “all non-generic word marks . . . eligible for protection so long as they satisfy
the ‘trademark use’ requirement”).
69
Bayer Co. v. United Drug Co., 272 F. 505, 509 (S.D.N.Y. 1921).
70
Of course, the question of the mark’s acquired distinctiveness—how well its
metaphor resonates in the marketplace—is another matter entirely. Such
distinctiveness may be relevant to how likely it is that consumers will assume that uses
of the mark by competitors are related to the mark holder.
71
Cf. PINKER, supra note 59, at 11 (“A name really has no definition in terms of
other words, concepts or pictures. Instead it points to an entity in the world, because
at some instant in time the entity was dubbed with the name and the name stuck.”).
Do Not Delete
1334
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
ranking “fanciful” marks stronger than “suggestive” ones, and so forth
down the line. Rather, the terms “fanciful” (or “fancy,” as some courts
called them) and “suggestive” were each used in particular cases to
distinguish the technical trademarks at issue in that case from merely
descriptive or generic words. These courts thus envisioned only a binary
distinction: marks that were distinctive because they functioned as the
proper name of a product or service and marks that merely described
72
some quality of the product.
The suggestion that a trademark’s validity depends not simply on
whether it operates on a metaphorical or naming level but also on how
metaphorical it is does, however, contribute to an understanding of
trademark ownership as the ability to control language use. First, the
Abercrombie hierarchy, by suggesting that marks that are more
metaphorical are entitled to a greater degree of protection, conveys an
expectation that control of all of an evocative mark’s subsequent uses and
associations will belong to the entity that developed the trademark. Some
courts have suggested, along these lines, that the greater protection given
to arbitrary and fanciful marks is justified because the trademark owner
has expended effort to create the metaphorical association among
consumers and thus deserves enhanced legal protection of the fruits of
73
its labor. Second, the hierarchy may contribute to increasing the scope
of the Lanham Act’s anti-dilution provision, which creates a cause of
action against uses of a famous and distinctive mark that are likely to
cause dilution of that mark, defined as “association arising from the
72
See, e.g., Le Blume Import Co., Inc. v. Coty, 293 F. 344, 351 (2d Cir. 1923)
(distinguishing between claimed trademarks that describe the characteristics of the
associated good and words that are “merely ‘suggestive’” of those characteristics);
Leonard v. White’s Golden Lubricator Co., 38 F. 922, 922 (C.C.S.D. Ohio 1889)
(rejecting defendant’s argument that “Valvoline” was compound of “valve” and
“oleine” (oil) and therefore descriptive, holding it to be fanciful); O’Rourke v.
Central City Soap Co., 26 F. 576, 578 (C.C.E.D. Mich. 1885) (holding that “antiwashboard” as applied to soap was not descriptive, “although the natural inference
from [the term] is that by the use of the soap the necessity of rubbing clothes is
obviated”). See also 2 MCCARTHY, supra note 23, at § 11:63 (stating that the “suggestive”
category arose in cases interpreting the 1905 Federal Trademark Act as a way of
“uphold[ing] the registration of marks that were only subtly ‘descriptive’”); id.
§ 11:12 (stating that courts “hardly ever” distinguish between “arbitrary” and
“suggestive” marks, since neither requires proof of secondary meaning).
73
See, e.g., Dreamwerks Prod. Grp., Inc. v. SKG Studio, 142 F.3d 1127, 1130 n.7
(9th Cir. 1998) (noting that with both fanciful and arbitrary marks, “the trademark
holder must work hard to make consumers associate the trademark with the product.
This suggests that any association is the result of goodwill and deserves broad
protection from potential infringers.”); BigStar Entm’t, Inc. v. Next Big Star, Inc., 105
F. Supp. 2d 185, 197 (S.D.N.Y. 2000) (“Recognizing the larger commitment of both
resourcefulness and resources that must be spent to procure market success with a
given suggestive term, the law rewards the extra demands by conferring broader
trademark protection upon this class than that granted to the merely descriptive
term.”). Relatedly, one court has suggested that descriptive or suggestive marks are
“better candidates for genericness” than fanciful trademarks. Ty Inc. v. Perryman, 306
F.3d 509, 513 (7th Cir. 2002).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1335
similarity between a mark or trade name and a famous mark” that is likely
either to impair the distinctiveness of the mark or to harm the reputation
74
of the mark. In essence, the anti-dilution statute allows mark holders to
control the metaphors associated with the famous mark, whether those
metaphors are distasteful or simply inconsistent with the mark holder’s
75
preferred associations. Despite the provision that the distinctiveness of
76
the famous mark may be inherent or acquired, one wonders whether
courts are more likely to find dilution the higher the mark falls on the
77
Abercrombie hierarchy; if that is the case, anti-dilution doctrine would
then tend to reinforce the notion that marks with stronger inherent
metaphorical associations can be used only in ways approved by their
creators.
This belief—that the only permissible types of metaphorical
trademark uses are the ones associated with the trademark by its
holder—does not, however, reflect the way in which consumers use
marks as parts of language. In particular, consumers, writers, and others
are often told that extending the metaphor associated with a trademark
to other, related concepts is an improper use of the mark and, in some
78
cases, risks adverse legal action. And yet, reasonable speakers not only
engage in such metaphorical wordplay often, but can do so without
79
losing sight of the mark’s original naming function. This is true not only
74
15 U.S.C. §§ 1125(c)(2)(B), (C) (2006).
Heymann, supra note 24, at 218–19 (“The brand owner, in [anti-dilution
cases], is claiming a right to the exclusive mental association with the brand in the
minds of the public.”); Sara Stadler Nelson, The Wages of Ubiquity in Trademark Law, 88
IOWA L. REV. 731, 735 (2003) (arguing that trademark owners who make similarly
dilutive uses of their marks should forego entitlement to a remedy under antidilution law).
76
15 U.S.C. § 1125(c)(1).
77
See, e.g., Nat’l Pork Bd. v. Supreme Lobster & Seafood Co., No. 91166701, 2010
TTAB LEXIS 225, at *63 (T.T.A.B. June 11, 2010) (“Even though inherent
distinctiveness is not required, the statute is weighted toward a finding of dilution
when the famous mark in question is commercially-strong and inherently distinctive.”
(citing Perfumebay.com Inc. v. eBay Inc., 506 F.3d 1165 (9th Cir. 2007))). The statute
provides that “[t]he degree of inherent or acquired distinctiveness of the famous
mark” can be considered by the court in determining whether the defendant’s mark
is likely to cause dilution of the plaintiff’s mark by blurring. 15 U.S.C.
§ 1125(c)(2)(B)(ii).
78
See NAOMI KLEIN, NO LOGO: TAKING AIM AT THE BRAND BULLIES 177 (1999)
(“[W]hen we try to communicate with each other by using the language of brands
and logos, we run the very real risk of getting sued.”); Rochelle Cooper Dreyfuss,
Expressive Genericity: Trademarks as Language in the Pepsi Generation, 65 NOTRE DAME L.
REV. 397 (1990); Alex Kozinski, Trademarks Unplugged, 68 N.Y.U. L. REV. 960, 975
(1993).
79
See, e.g., Dreamwerks Prod. Grp., Inc. v. SKG Studio, 142 F.3d 1127, 1132 (9th
Cir. 1998) (“If you received junk e-mail, you were Spammed. The childish antics of
politicians are Mickey Mouse. Lousy, mindless work is a McJob. A quick fix is a BandAid. Glitz and ditz make for a Barbie World. Calling something the Rolls Royce of its
class is shorthand for referring to a refined product targeted at those with expensive
tastes. And maudlin family gatherings make for Kodak moments. None of these
75
Do Not Delete
1336
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
of well-documented uses such as “putting a Band-Aid on the situation” or
calling something the “Cadillac of its class” but also of more playful,
consumer-driven metaphorical uses. For example, the “Mc-” formative of
80
McDonald’s has become a generative metaphor, spawning not only
references to a “McJob” or a “McMansion,” the former of which was
81
added to Merriam-Webster’s Dictionary in 2003, but consumer-created
brand extensions, such as the “Mc10:35,” a combination of a double
hamburger and an Egg McMuffin, so named because 10:35 a.m. is the
most likely time to be able to order both a breakfast item and a lunch
82
item at a McDonald’s restaurant.
Trademark holders often bristle at these uses, sending cease-and83
desist letters or making other entreaties to try to curb such neologisms.
The argument, apparently, is that the extension of the metaphor from
corporate-created associations to consumer-created associations causes a
weakening of the brand, such that the original metaphor no longer
persists—that, for example, the prevalence of “McMansion” means that
84
“McDonald’s” will cease to refer to the fast-food restaurant chain. But
there is no reason to think that this will be the case. Many users of the
term will be well aware that the meaning of “McMansion” derives from
the extension of the metaphors associated with the McDonald’s
trademark (mass-production and uniform design) to single-family-home
architecture. Such consumers will understand that a newspaper’s
phrases had any meaning before the trademark was absorbed into the language; each
is evidence of how commerce and culture transform each other.”); see also Lucasfilm,
Ltd. v. High Frontier, 622 F. Supp. 931, 935 (D.D.C. 1985) (“The new meaning of the
phrase in the political or scientific context does not affect the distinct, and still strong
secondary meaning of STAR WARS in trade and entertainment.”) (rejecting attempt
to enjoin use of “Star Wars” as a popular name for Ronald Reagan’s strategic defense
initiative); Gerhardt, supra note 49, at 487 (arguing for a “cultural dilution” exception
for such uses).
80
PINKER, supra note 59, at 240 (describing generative metaphors).
81
Rupert Cornwell, Dictionary Definition of ‘McJob’ Is Slap in Face, Says Angry Burger
Boss, THE INDEPENDENT (U.K.), Nov. 10, 2003, http://www.independent.co.uk/news
/world/americas/dictionary-definition-of-mcjob-is-slap-in-face-says-angry-burger-boss735176.html.
82
Phil Villarreal, Check Out the Secret McMenu Item That’s Sweeping SF, the Mc10:35,
THE CONSUMERIST (Mar. 5, 2010, 8:46 AM), http://consumerist.com/2010/03/checkout-the-secret-mcmenu-item-thats-sweeping-sf-the-mc1035.html. See also SHUY, supra
note 17, at 102–03 (asserting that media uses of “Mc” (as in “McJob”) involve
metaphorical meanings).
83
See, e.g., Cornwell, supra note 81 (describing the opposition of McDonald’s to
dictionary entry).
84
See SHUY, supra note 17, at 103 (contending that such uses indicate that the
prefix has “come [to] develop its own meaning” and has “broken loose from
association with McDonald’s”). Of course, this example does not follow directly from
the Abercrombie hierarchy’s influence, since “McDonald’s” is not an invented word.
But, I contend, the propertization of trademarks that is embedded in Abercrombie’s
language leads naturally to a view of trademarks and their associations as inviolate.
What is characterized as a fear of genericness, therefore, may really be a resistance to
undesirable associations.
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1337
reference to “McMansions” is extending the underlying metaphor in a
playful manner and not describing a new offering from the restaurant
chain. And even if other users do not make the connection, this does not
mean that such users are likely to consider “McDonald’s” generic for
85
hamburger restaurants. To be sure, extensions that use the metaphor in
a way that causes consumers to think that McDonald’s is responsible for
86
the good or service at issue constitute infringement. But the use of the
metaphorical association in other, expressive contexts would not have
meaning for many if the metaphor were not robust.
In sum, courts should be aware of the atmospheric effects that the
Abercrombie spectrum of distinctiveness can have. By encouraging
trademark holders and others to think of trademarks in terms of degrees
of strength, rather than in terms of metaphor, the hierarchy suggests that
the more metaphorical a mark is or has become, the less it becomes
available for additional metaphorical use. This is not to say that judges
can no longer rely on their intuition about the metaphorical function of
a mark; such a function may be assumed from the type of mark at issue or
87
its context, such that survey or other evidence is not required. It is to
say, rather, that how the concept of trademark validity is framed is
important. Thinking of certain marks as better or stronger than others
suggests a concomitant expansion of dominion over those trademarks
that can carry over to the noncommercial context. This mindset then
encourages trademark holders to promulgate additional rules in an
attempt to further control how those marks are used. As I now describe,
however, perfectly reasonable communication about trademarks can take
place in violation of these rules, without any indication that consumers
have lost sight of the status of the trademarks at issue.
B. Rule #2: Trademarks Should Not Appear in Lowercase
Wordplay is an important part of one’s ability to communicate in a
language. To be conversant in parody, puns, and other literary tropes is
to fully participate in popular culture. Even more mundane
communications require a certain agility in comprehension—for
example, to continue the discussion on metaphor, a reasonable speaker
of English would feel comfortable saying that something was “water
85
Likewise, the fact that a famous individual may lend his or her name to a
metaphorical use does not prevent that name from continuing to identify the
individual, even if some speakers are unaware of the derivation. See infra notes 92–94
and accompanying text.
86
See, e.g., McDonald’s Corp. v. Druck & Gerner, DDS., P.C., 814 F. Supp. 1127,
1135 (N.D.N.Y 1993) (enjoining use of “McDental” for dental practice).
87
Cf. Dinwoodie, supra note 13, at 143 (noting that many of trademark law’s
rules “have been induced over time from the reasoned outcomes in individual cases,
where the outcome to some extent has been informed by how language has been
used and understood in the marketplace”). On the relevance of administrative costs
to such decision making, see generally Robert G. Bone, Enforcement Costs and
Trademark Puzzles, 90 VA. L. REV. 2099 (2004).
Do Not Delete
1338
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
under the bridge” without worrying that her listener would think she was
engaging in a discussion about civil engineering. Slang, in particular, is
organically formed, taking on meaning through shared understanding.
The first person to use “cool” to mean “well-regarded” (as in, “That new
88
song is really cool”) may have been met with blank stares; today, no one
either questions this use of the word in context or finds herself without a
word to describe the temperature that is the opposite of warm. Context is
the key to understanding which meaning of “cool” is intended.
Indeed, as linguists have recounted, most of the words we use in the
English language are polysemic—that is, they are associated with several
89
(related) meanings. Thus, it is rather pointless to ask what a word
means because the answer depends on the context in which the word is
90
used and the community within which one is speaking and listening.
This fact does not, of course, render a word incapable of carrying
meaning; indeed, a single word can be associated with two facially
contradictory meanings without losing its legitimacy as a word in the
English language. Consider, for example, the word “dust,” which can
mean both to put a powder on something (“to dust a cake with powdered
sugar”) and to remove a powder from something (“to dust the
bookshelves”), or the word “trim,” which can mean both to remove
something (“to trim one’s hair”) and to add something (“to trim a
91
Christmas tree”).
The same phenomenon exists once one moves into naming
practices, even though here one of the links is to an association rather
than to a meaning. Although capitalization will often distinguish a
proper name from a common usage, this is not always the case,
particularly in oral communication. Yet, speakers typically have few
problems negotiating the various associations attached to a single lexical
unit. One can say of another that he has a Napoleon complex without
92
losing sight of the fact that “Napoleon” also refers to a historical figure;
the nonalcoholic cocktail made with ginger ale and grenadine and often
enjoyed by children is a Shirley Temple, as is the child film star and later
88
For a discussion of the etymology of this meaning, see Ben Zimmer, Cool, N.Y.
TIMES, May 30, 2010, § E (Magazine).
89
Polysemy, in which a word’s multiple meanings are related, is often
distinguished from homonymy, in which a word’s multiple meanings are unrelated,
although the line can be difficult to draw. ALLAN, supra note 27, at 147–48. However
denominated, the discussion is not completely distinct from the discussion of
metaphor above.
90
CRYSTAL, supra note 33, at 191 (“The word charge could be something to do
with money, crime, soldiering, electricity, or explosives—to take just five
possibilities.”); Geoffrey Nunberg, The Non-Uniqueness of Semantic Solutions: Polysemy, 3
LINGUISTICS & PHIL. 143, 175–76 (1979) (suggesting that language is a matter of
collective agreement).
91
See, e.g., Clark & Clark, supra note 36, at 793.
92
See LANDAU, supra note 38, at 212 (describing how such words, which develop
“one or more senses distinct from any necessary relation with the person after whom
the word was formed,” will usually have a separate definition in a dictionary).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1339
U.S. representative to the United Nations and ambassador to Ghana for
93
whom the drink was named. In neither of these instances does the use
in one area require relinquishing the reference in another area; indeed,
it is likely that at least some users of these terms are unaware of the
94
individuals from whom the term was derived. Linguistic theory advises
us that these types of uses do not involve, technically, the same word—
rather, they involve, like other names, separate units occupying the same
95
lexical space. This is true even when the same lexical unit appears both
in capitalized form (to indicate a proper name) and in lowercase form
(to indicate a common noun). The concept of “faith,” for example, may
be spelled in the same way as the name “Faith,” but that doesn’t make the
two the same “word”—one is a common noun that can be defined as
“trust in something without proof,” and one is a reference to a female
individual, with no definition beyond its reference. The exhortation “I
have faith, and that’s all I need” means one thing when uttered by a
congregant; “I have Faith, and that’s all I need” means something
different when uttered by Faith’s partner. The existence of the common
noun does not, in and of itself, weaken the ability of the proper noun to
serve as a name, nor is qualification of the proper noun typically
96
necessary to clarify its meaning.
The concept of polysemy is already inherent in trademark law in at
least two respects. First, trademark law accommodates the practice of
allowing a lexical unit to serve as the name for two different kinds of
products at the same time, such as “Delta” for airlines and “Delta” for
faucets. Such marks are allowed to be used by different producers for
distinct products or services because they operate in different markets
93
EUGENE EHRLICH, WHAT’S IN A NAME?: HOW PROPER NAMES BECAME EVERYDAY
WORDS 204 (1999). A Roy Rogers, named after the television cowboy, is a similar
nonalcoholic drink typically served to children. Id.
94
Such terms sometimes undergo a period of transition, during which
knowledge of the original naming association is essential to understanding meaning.
See, e.g., Clark & Clark, supra note 36, at 784 (explaining how the use of Houdini in
sentences such as “My sister Houdini’d her way out of the locked closet” depends
“not merely on one’s knowledge of English, but also on one’s knowledge of particular
facts about Houdini the historical figure”).
95
Cf. Pullum, supra note 25 (“Jack Daniel’s is a product (a bourbon), and not a
trademark; Jack Daniel’s is a trademark—you can’t serve it on the rocks.”).
96
The language of some courts has reflected a misunderstanding of this
distinction in trademark cases. See, e.g., Sullivan v. CBS Corp., No. 00 C 5060, 2002
U.S. Dist. LEXIS 6625, at *15–16 (N.D. Ill. Apr. 8, 2002), aff’d, 385 F.3d 772 (7th Cir.
2004) (stating, in a case involving “Survivor” the television show and “Survivor” the
rock band, that “[i]n fact, the word ‘survivor’ is a commonly used noun that appears
in a standard dictionary in lower case. As such, it could be deemed generic and thus
unworthy of any trademark protection.” (dictum)); Tsiolis v. Interscope Records,
Inc., 946 F. Supp. 1344, 1352 (N.D. Ill. 1996) (holding, despite federal registration
for “Aftermath” as the name of a heavy metal band, that the word “aftermath” was “a
term undeserving of trademark protection” because “the noun ‘aftermath’ appears
entirely in lower-case letters” in two dictionaries and had no secondary meaning
relating to band).
Do Not Delete
1340
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
97
and so are unlikely to give rise to confusion among consumers.
Notwithstanding this characteristic, it is likely that consumers will at some
point encounter such marks without any accompanying generic word
(“I’m flying Delta to my brother’s wedding”). Trademark law thus
assumes that consumers have the agility to provide the appropriate
association given context; the fact that the same word, absent context,
could have more than one association does not deprive the word of
98
trademark status. More perniciously, perhaps, trademarks have been
adopted as drug slang, with “Lucky Charms,” “M&M’S,” and “Kleenex” all
99
used as words for ecstasy. One does not anticipate, however, that
consumers of such products are unable to buy cereal, chocolate, and
facial tissues when they go to the supermarket; the only confusion may be
among law enforcement officers who are not conversant in the relevant
language.
Second, as discussed earlier, trademark protection is not limited to
invented words. Trademark law provides that a lexical unit can serve both
as the name of a product or service and as a common word with a non100
trademark meaning, such as “Apple” for consumer electronics and
“apple” for a piece of fruit, and reasonable consumers recognize that
these multiple references can share the same lexical space without
rendering any reference nugatory. As with other types of words,
context—a stylized font or the sentence structure of advertising copy, for
example—allows the consumer to select from among various linguistic
options. Trademark law incorporates this concept most directly with
respect to so-called descriptive marks—marks that are co-existent with
words that can describe a quality or characteristic of the property. Once
such marks develop “secondary meaning,” an association that signifies
the word as a trademark, the lexical unit now serves two functions: one as
the common word, which any manufacturer should be free to use to
101
and one as the trademark. But the same
describe its product,
phenomenon takes place with any trademark that isn’t an invented word.
The use of “Camel” for cigarettes similarly creates two lexical units
sharing the same word space: one that functions as the proper name for a
97
See, e.g., Dawn Donut Co., Inc. v. Hart’s Food Stores, Inc., 267 F.2d 358, 364
(2d Cir. 1959).
98
See, e.g., Durant, supra note 15, at 109; Heymann, supra note 24, at 220;
Rebecca Tushnet, Gone in Sixty Milliseconds: Trademark Law and Cognitive Science, 86
TEX. L. REV. 507, 529–30 (2008).
99
MICHAEL ADAMS, SLANG: THE PEOPLE’S POETRY 39 (2009) (noting the “clever
placement of ecstasy’s first syllable as the final syllable of Kleenex, where no one
expects a drug reference”).
100
Jessica Litman, The Public Domain, 39 EMORY L.J. 965, 973 (1990) (noting that
“ownership of a trademark gives rise to what is, in essence, a homonym”).
101
See 15 U.S.C. § 1115(b)(4) (2006) (providing a defense to infringement to a
party whose “use of the name, term, or device charged to be an infringement is a use,
otherwise than as a mark, . . . of a term or device which is descriptive of and used
fairly and in good faith only to describe the goods or services of such party, or their
geographic origin”).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1341
particular type of cigarette and one that identifies a type of animal.
Trademark law thus assumes that consumers will become comfortable
with the fact that many, if not most, of the words claimed as trademarks
will also be used in connection with other meanings and associations in
other contexts, with capitalization sometimes, but not always, providing a
102
key to interpretation.
Yet despite these assumptions, trademark holders and some
commentators assert that any time a word used as a trademark appears in
lowercase type, consumers are unlikely to be able to engage in such
interpretive efforts, such that the validity of the mark will be chipped
103
away. This belief is rehearsed by scholars, by associations such as the
International Trademark Association, which promulgates media guides
104
and even by word
delineating the proper uses of trademarks,
processing programs such as Microsoft Word, which will autocorrect
certain words that share a lexical unit with trademarks by replacing a
lowercase letter with a capital, in something akin to a reverse Cupertino
105
effect. Whether a trademarked term appears in lowercase may not be
irrelevant to determining whether speakers view the word as continuing
106
to serve as the proper name of a product or service. But context is
critically important to distinguishing between lowercased uses that
suggest that consumers no longer recognize the proper name as such
and lowercased uses that do not so suggest. For example, as has been
well-documented, in the southern portion of the United States, “Coke” is
102
Rochelle Cooper Dreyfuss, We Are Symbols and Inhabit Symbols, So Should We Be
Paying Rent?: Deconstructing the Lanham Act and Rights of Publicity, 20 COLUM.-VLA J.L. &
ARTS 123, 149 (1996) (“If, indeed, symbols are routinely assigned multiple meanings,
the law ought to start from the premise that audiences have a great deal of
experience with confusing signals and that, absent a specific reason to believe
otherwise, the interpretive strategies that a listener ordinarily pursues will also be
used in the marketplace.”). A linguist might refer to the distinction conferred by
capitalization as an example of prosody. See, e.g., 2 ALLAN, supra note 27, at 1–2.
103
See, e.g., John Dwight Ingram, The Genericide of Trademarks, 2 BUFF. INTELL.
PROP. L.J. 154, 160 (2004).
104
INT’L TRADEMARK ASS’N, A GUIDE TO PROPER TRADEMARK USE FOR THE MEDIA
AND PUBLISHING PROFESSIONALS (2009), http://www.inta.org/downloads/documents
/2009_TrademarkUse.pdf.
105
The term “Cupertino effect,” which refers to the occasion when a computer’s
spell-checker automatically replaces an unknown word with an undesired word,
originated from the substitution of the word “Cupertino” for the words “cooperation” or “cooperation,” resulting in references such as “Cupertino agreements”
in European Union documents. Elizabeth Anne Muller, Cupertino and After,
LANGUAGE MATTERS, Sept. 2000, at 9, 10; Benjamin Zimmer, The Cupertino Effect,
LANGUAGE LOG (Mar. 9, 2006, 4:21 PM), http://itre.cis.upenn.edu/~myl/
languagelog/archives/002911.html.
106
See, e.g., Door Sys., Inc. v. Pro-Line Door Sys., Inc., 83 F.3d 169, 171 (7th Cir.
1996) (“If the term ‘door systems’ appeared in a standard dictionary in lower case,
this would be powerful evidence that the term was generic, because nouns and other
nominatives listed in dictionaries, save for the occasional proper name, denote kinds
rather than specific entities (‘dog,’ not ‘Fido’).”).
Do Not Delete
1342
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
used to refer to the soft drink manufactured by the Coca-Cola Company,
107
while “coke” is used to refer to any soft drink. Thus, it would not be
unusual for a restaurant patron to respond to the question “What kind of
108
coke do you want?” with “I’d like a Coke coke, please.” The use of the
lowercased (by implication in aural speech) “coke” does not indicate that
“Coke” no longer serves as a proper name; rather, it simply means that
the lowercased form is used to refer to a category of similar items
(“drinks that are like Coke”), just as “xerox” is often used to refer to a
category of photocopiers, of which “Xerox” is one, and “tylenol” is used
to refer to the drug with the formulaic name of acetaminophen, of which
109
“Tylenol” is a particular brand name. Indeed, it may well be the case
that, in some instances, use of the lowercased form of the trademark
involves fewer communication costs than the corresponding common
noun. For example, some consumers searching for competitors to
Tylenol may not be aware of the common noun “acetaminophen”; using
“tylenol” to mean “pain relievers that are formulated like Tylenol” helps
to ensure that such consumers get what they want without imposing the
110
burden of learning pharmacological vocabulary.
Courts that purport to take account of the reasonable consumer’s
perceptions and beliefs should therefore recognize that the appearance
of a trademark in lowercase type does not, of itself, indicate that the word
no longer functions as a proper name in the appropriate context. More
specifically, courts should be open to taking a nuanced approach,
recognizing that speakers of a language often use the same lexical unit in
association with more than one reference and rely on context to supply
107
Scott Leith, In the South, a “Coke” Could Be a Pepsi: Pop, Anyone? Soda? Generic
Term for Drinks Is Regional, ATLANTA J. CONST., Jan 27, 2005, at 1E (describing study
tracking regional variations in term used for soft drinks).
108
Adams, supra note 49, at 7 (criticizing dictionary publisher for failing to
recognize the difference between “Coke” as a trademark and “coke” as a southern
regionalism for a soft drink).
109
As Steven Pinker has highlighted, a small lexical difference can suggest a
distinction between a particular item and a category of items. PINKER, supra note 59, at
165 (noting that in Google’s keyword buy program, plurals (such as “digital
cameras”) cost more than singulars (“digital camera”) because the singular “is
generic, and is likely to be typed by someone who wants to know how they work,”
while a plural “is more likely to be referential, and typed by someone who wants to
know about the kinds that are out there and how to get one”). See also, e.g., Desai &
Rierson, supra note 47, at 1847 (noting that someone asking a colleague for a
“kleenex” may not care what brand of tissue he is given, but someone buying tissue in
the supermarket may want to buy Kleenex to the exclusion of other brands); Michael
Grynberg, Things Are Worse Than We Think: Trademark Defenses in a “Formalist” Age, 24
BERKELEY TECH. L.J. 897, 920 (2009) (noting that a generic drug manufacturer’s use
of “Tylenol” as a search engine keyword “may serve as the junior user’s signal that ‘I
am in the same product category as TYLENOL,’ not ‘I am TYLENOL’”).
110
Simonson, supra note 28, at 194 (“In many cases, buyers use trademarks as
generic terms, not only casually and for convenience purposes, but because they
apparently do not know other effective ways to communicate the type of product they
seek.”).
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1343
any necessary clarification. A similar distinction was recognized by the
111
court in the well-known 1921 case of Bayer Co. v. United Drug Co., in
which the question was whether the term “aspirin” had become generic.
Using an approach that would likely not find favor among courts today,
the court deemed the term generic among consumers but a trademark
(as “Aspirin”) with respect to manufacturing chemists, druggists, and
physicians, noting that if trademark law could allow different uses of the
same word among “vertical divisions of the demand,” the same “must
112
apply to horizontal.” While the remedy may have been unique, the
underlying principle still holds: The ultimate question is what “buyers
113
understand by the word for whose use the parties are contending.” The
use of a lowercase form does not, by itself, confirm that a trademark has
become generic absent an indication that a capitalized use no longer
refers to the trademarked good or service.
C. Rule #3: A Trademark Must Always Be Used as an Adjective, Never as a
Noun (Including Possessive or Plural Forms) or as a Verb
Trademark owners and commentators will often suggest that a
trademark must always be used as an adjective modifying a generic
common noun, such as in the phrase “Ford car” or “Xerox
114
photocopier.” (It may be that the “trademarks must be adjectives” rule
comes from the fact that words used as descriptive marks are typically
111
272 F. 505 (D.N.Y. 1921).
Id. at 514. See also, e.g., Am. Thermos Prods. Co. v. Aladdin Indus., Inc., 207 F.
Supp. 9, 27 (D. Conn. 1962) (granting defendant the right to use “thermos” so long
as it was preceded by “Aladdin’s” and reserving to plaintiff the right to use the
capitalized “Thermos,” stating “[t]herefore, those consumers who desire to buy the
trademarked products of the plaintiff can identify them by exactly the same
trademarks that they have always seen”).
113
Bayer, 272 F. at 509. See also, e.g., E. I. DuPont de Nemours & Co. v. Yoshida
Int’l, Inc., 393 F. Supp. 502, 524 (E.D.N.Y. 1975) (“It cannot be thought unusual for a
generic term describing the nature or class of an article . . . to be associated with its
major indicator of source . . . but it does not necessarily follow that the public has lost
sight of the distinction.”); GEOFFREY K. PULLUM, Punctuation and Human Freedom, in
THE GREAT ESKIMO VOCABULARY HOAX AND OTHER IRREVERENT ESSAYS ON THE STUDY OF
LANGUAGE 67, 73 (1991) (contending that “gratuitous capitalization destroys
distinctions: consider the contrast between the words Xerox (the name of a
corporation) and xerox (a verb that the Xerox Corporation has tried desperately to
eliminate, and forbids its staff to use)”).
114
See, e.g., 1 ANNE GILSON LALONDE, GILSON ON TRADEMARKS § 2.02[6] (Karin
Green ed., 2010) (providing practice tips to reduce risk of genericism, including
using word marks as “a proper adjective that modifies a noun” and not using a
trademark in the possessive, in the plural, or as a verb); Sidney A. Diamond, How to
Use a Trademark Properly, 61 TRADEMARK REP. 431, 432, 438–39 (1971) (discussing rules
for proper trademark use, including ensuring use as a “proper adjective” and
avoiding “incorrect grammatical form[s],” plurals, and possessives); INT’L TRADEMARK
ASS’N, supra note 104 (noting that “Please hand me a KLEENEX tissue” is correct,
while “[It was] a five-KLEENEX movie” is incorrect).
112
Do Not Delete
12/15/2010 10:02 PM
1344
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
115
adjectival, but it is probably more accurate to characterize trademarks
116
as proper names and, therefore, more like nouns. ) To use a trademark
as a noun—as in, “I’m going to buy a Ford next week,” “My Ford’s
interior is black,” or “There are three Fords in the parking lot” or as a
verb—“Could you please xerox that document for me?”—is seen as
evidence that the mark has lost its naming function and become generic.
Courts, for their part, can be imprecise about the effect of such uses,
suggesting that they are strong evidence that a proper name has lost its
117
trademark function.
But the literary device known as “anthimeria,” in which one part of
speech is substituted for another, dates back at least as far as
Shakespeare, as Arthur Quinn has noted: “Thank me no thankings, nor
proud me no prouds,” from Romeo and Juliet, and “Every why hath a
118
wherefore” from Comedy of Errors are but two examples. As with
115
Nartron Corp. v. STMicroelectronics, Inc., 305 F.3d 397, 404 (6th Cir. 2002)
(stating that a “merely descriptive” mark is “very similar to an adjective” in that it “is
often said to identify a characteristic of the thing”); Royal Baking Powder Co. v.
Sherrell, 93 N.Y. 331, 334 (1883) (“Letters or figures which, by the custom of traders
or the declaration of the manufacturers, are only used to denote quality, are
incapable of exclusive appropriation, but are open to use by any one, like the
adjectives of the language.”).
116
Cf. Folsom & Teply, supra note 28, at 1325 n.21 (1980) (“The disagreement
over whether a trademark is a proper adjective or a proper noun reflects the
possibility that a trademark can be used elliptically, omitting the generic term when it
is obviously understood.”); Pullum, supra note 25 (contending that trademarks are
used as “attributive modifiers,” not as adjectives).
117
See, e.g., FreecycleSunnyvale v. Freecycle Network, Inc., No. C 06-00324 CW,
2006 U.S. Dist. LEXIS 74767, at *8–9 (N.D. Cal. Oct. 3, 2006) (holding that
counterclaim that plaintiff had “willfully and intentionally induced third parties to
infringe the Marks by encouraging others to misuse the Marks in the form of verbs,
adjectives, gerunds, and participles . . . so that this misuse will result in the Marks
being rendered generic” stated a cause of action for contributory infringement);
Birtcher Electro Med. Sys., Inc. v. Beacon Labs., Inc., 738 F. Supp. 417, 420 (D. Colo.
1990) (“[Defendant] showed numerous instances in which [Plaintiff’s] literature and
videotaped advertising used the terms [at issue] as nouns as opposed to adjectives,
indicating that the terms were generic labels for the product itself.”). Some courts
that appear to be favorable to uses such as “I’ll have a Budweiser” err in describing
this use as “generic,” albeit a usage that does not render the mark invalid. See, e.g.,
Caesars World, Inc. v. Milanian, 247 F. Supp. 2d 1171, 1194–95 (D. Nev. 2003)
(“[O]ccasional generic usage does not destroy a trademark. Indeed, often the use of
a trademark as a generic noun illustrates the power of a trademark. . . . [T]he
consumer ordering ‘a hamburger and a Coke’ is demonstrating brand loyalty, not
genericness.”), aff’d, Nos. 03-16063, 03-16685, 2005 U.S. App. LEXIS 1533 (9th Cir.
Feb. 1, 2005). Some courts, however, indicate an understanding of the distinction.
See, e.g., Feathercombs, Inc. v. Solo Products Corp., 306 F.2d 251, 256 (2d Cir. 1962)
(“A mark is not generic merely because it has some significance to the public as an
indication of the nature or class of an article. If that were so, then Feathercombs’
frequent reference to its product as ‘the Feathercomb’ might well have caused it to
lose exclusive right to the [mark].”).
118
ARTHUR QUINN, FIGURES OF SPEECH: 60 WAYS TO TURN A PHRASE 50–51 (1982)
(citing WILLIAM SHAKESPEARE, ROMEO AND JULIET, act 3, sc. 5, and WILLIAM
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1345
metaphors, anthimeria rely on their contraposition to an existing
structure for the conveyance of meaning; hence, it would be illogical to
assume that the structure is eviscerated simply because the device is
119
employed. Thus, the claim that the only correct use of a trademark is as
an adjective modifying a generic noun rarely reflects how reasonable
consumers tend to speak. Conversations about trademarks often include
a form of anthimeria; for example, it is common for speakers to use a
trademark in a noun form (such as “I’d like a Budweiser with my lunch,
please”), with no indication that they fail to get the products they
120
desire. And once a trademark is used as a noun, it would not be
surprising, from a linguist’s perspective, that possessive and plural uses
121
would follow. Indeed, at least one court has recognized its own
tendency to use trademarks in this way. In American Dairy Queen Corp. v.
122
New Line Productions, Inc., the plaintiff, who held a trademark in “Dairy
Queen” for a chain of ice-cream restaurants, sought to enjoin the
producers of a mock documentary—about beauty contests in rural
Minnesota—who planned to call the film “Dairy Queens,” a wordplay on
123
the dairy industry in that part of the country. The plaintiff claimed that
the use of “Dairy Queens” for the title of the film would “cause the public
to associate its trademarked name with the unwholesome content of the
124
film.” In concluding that the plaintiff was likely to achieve success on
SHAKESPEARE, COMEDY OF ERRORS, act 2, sc. 2). Linguists might call this device
“conversion” or a “functional shift.” See, e.g., CRYSTAL, supra note 33, at 227 (providing
“to referee” as an example of a noun-to-verb conversion).
119
QUINN, supra note 118, at 52 (“The anthimeria, to be an anthimeria, needs a
stable grammatical structure to rebel against. Only a theist can be a sinner.”).
120
GILSON LALONDE, supra note 114, § 2.02[6] (noting that, for certain products,
“manufacturers and consumers use the mark as a noun”); 1 MCCARTHY, supra note 23,
§ 12:27 (“The person who orders for lunch a ‘BIG MAC and a COKE’ undoubtedly
has brand knowledge and brand loyalty. . . . Since everyone knows the generic names,
they are dropped in ordinary usage.”); id. § 12:8 (noting that uses such as “[m]ake
me a Xerox of this” are “casual, non-purchasing uses” that constitute “very weak
evidence of generic usage”); Ronald R. Butters, A Linguistic Look at Trademark Dilution,
24 SANTA CLARA COMPUTER & HIGH TECH. L.J. 507, 514 (2008) (noting that consumers
can recognize a mark as a brand name “even if consumers sometimes use the brand
name figuratively or in shorthand (i.e., synecdochical) reference to the type of goods
or service that each brand represents”); Diamond, supra note 114, at 443–44
(acknowledging that consumers often use trademarks without an accompanying
generic term). Of course, additional information can be provided to resolve any
lingering interpretation issues, such as when a restaurant server responds to the
request for a “coke” by saying, “We serve only Pepsi. Is that OK?”
121
See, e.g., Pullum, supra note 25 (“[S]ince every regular noun has a genitive
form, every trademark that has the form of a singular noun has a genitive form too:
My Porsche’s top speed is 130 mph is not rationally regarded as a misuse of a
trademark.”).
122
35 F. Supp. 2d 727 (D. Minn. 1998).
123
The film did not reference the restaurants or ice cream in any way. Id. at 729.
After losing the case, New Line released the film under the title “Drop Dead
Gorgeous.”
124
Id.
Do Not Delete
1346
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
the merits (a questionable result), the court noted that the two marks
were “identical—save and except the single letter ‘s’ which makes New
125
Line’s term plural.” The pluralization of the Dairy Queen mark, the
court noted, was irrelevant, since “Dairy Queens” was likely to be
understood as the equivalent of “Dairy Queen,” particularly in light of
the court’s own experience in “taking its children for ‘Dairy Queens’ on a
126
warm summer’s evening.”
Indeed, in some cases, the way in which trademarks are rendered in
plural form bolsters their status as proper names rather than as common
nouns. Because trademarks are, like other names, “stretches of sound” as
opposed to embodiments of the “canonical roots” of the words with
which they share a lexical unit, their plural forms are often constructed
differently from the way in which they would be formed if they were not
127
proper names. Thus, just as Julia Child and her husband, Paul, would
be referred to jointly as the Childs, not the Children, the plural of
“Mickey Mouse” is usually “Mickey Mouses,” not “Mickey Mice”; the
plural of “Sony Walkman” is typically not “Sony Walkmen”; and the
128
Toronto hockey team is the Maple Leafs, not the Maple Leaves. In each
case, the root that appears to be included within each mark (“mouse,”
“man,” and “leaf”) in fact does not exist; the identical lexical unit serves,
in the trademark, solely as a name and not as a common noun.
Reasonable speakers of English also regularly engage in the literary
trope of metonymy, another type of noun use. Metonymy refers to the
use of one name to refer to a larger entity, such as the use of “the White
House” in the phrase “the White House issued a statement today” to refer
to the U.S. government or “Wall Street” to refer to New York’s financial
industry. (Some might refer to this particular type of metonymy as
synecdoche, which is the use of a part of something to refer to the whole,
such as the phrase “hungry mouths to feed.”) Trademarks are likewise
used as a substitute for the corporation itself or one of its employees,
such as the use of “Budweiser” as a substitute for “the beer made by the
Anheuser-Busch company” (as in “I’d like a Budweiser, please”) or the
use of “the Times” in the phrase “the Times hasn’t arrived at the press
conference yet” to indicate a reference to “the reporter from the
129
Times.” Trademarks can also be used to refer to the owners of the
125
Id. at 730.
Id.
127
PINKER, supra note 35, at 153–56 (noting that onomatopoeia works in this way,
such that the past tense of “beep” is “beeped,” not “bept”).
128
Id. at 163 (noting that because “Walkman” doesn’t have the word “man” as its
root, the plural is usually “Walkmans” rather than “Walkmen”); id. at 164 (explaining
that the plural is typically “Mickey Mouses” because “Mickey Mouse” as an expression
no longer involves reference to an animal).
129
LAKOFF & JOHNSON, supra note 32, at 35. As Steven Pinker points out,
polysemic associations can be combined in the same expression. PINKER, supra note
59, at 110 (giving as an example “The Boston Globe decided to change its size and
126
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1347
products with the marks or to the consumers who buy the products. At a
service station, for example, the manager might tell a mechanic, “Go out
and tell the blue Volkswagen that her car is ready,” while a restaurant
130
server might tell a colleague, “The Dr. Pepper is sitting at table six.”
None of these uses follows the purported rules of trademark grammar—
indeed, each expression on its face uses the mark to refer to something
other than the trademarked good—and yet the trademark function of
the term is unchanged.
Use of a trademark as a verb, such as “I’m going to xerox that
document,” “Did you remember to tivo that television program?” or “I
need to photoshop my wrinkles out of that photograph,” is equally
common. In some cases, the verbs involve the use only of the
trademarked good rather than a similar product in its class. “I’m going to
google the answer to that” probably still involves the use of the Google
search engine; one does not (yet) google any information on Yahoo! or
“greyhound” to any city except on a Greyhound bus. In other cases, the
trademark-as-verb may refer to an action accomplished not with the
trademarked product but with another product in its class. Speakers use
“xerox” to refer to making a copy of a document on any photocopier,
whether a Xerox photocopier or not, and “photoshop” to refer to the act
of digital photo editing generally, not just by using Adobe’s Photoshop
131
software. As linguists have noted, this conversion may be motivated by a
conversational vacuum. In areas of new technology, speakers of a
language reach for a verb form to convey the action they take when they
make use of a new type of product. An innovative product, or the first of
its kind, fills this gap—hence, the verb “to xerox” might have been
typeface,” in which “Boston Globe” refers both to the company and to the physical
newspaper it produces).
130
Cf. PINKER, supra note 59, at 112.
131
Indeed, courts have used the terms “xeroxed” or “Xeroxed” as a synonym for
“photocopied” in a multitude of cases, including a handful of intellectual property
cases. See, e.g., Silverstein v. Penguin Putnam, Inc., 522 F. Supp. 2d 579, 597 (S.D.N.Y.
2007) (“To prepare the manuscript for Complete Poems, Ms. Breese cut and pasted or
Xeroxed sheets directly from previous Parker compilations.”). One Eleventh Circuit
opinion originally contained such a usage; the opinion was reissued with the term
revised. Compare St. Charles Mfg. Co. v. Mercer, 719 F.2d 380, 381 (11th Cir. 1983)
(case involving “St. Charles” trademark for kitchen cabinets) (“By this time, the
majority of brochures (approximately 3000) had been distributed. Consequently,
defendants made the corrections primarily on xeroxed copies.” (emphasis added)), with
St. Charles Mfg. Co. v. Mercer, 737 F.2d 891, 892 (11th Cir. 1983) (“By this time, the
majority of brochures (approximately 3000) had been distributed. Consequently,
defendants made the corrections primarily on photocopies.” (emphasis added)).
Similarly, “photoshopped” appears in more than one opinion. See, e.g., Furminator,
Inc. v. Kim Laube & Co., Inc., No. 4:08CV00367 ERW, 2009 U.S. Dist. LEXIS 118930,
at *9 n.1 (E.D. Mo. Dec. 21, 2009) (describing a photograph that had been
“photoshopped” to include a false date in an attempt to create prior art in a patent
case).
Do Not Delete
1348
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
needed when the technology was fairly new, whereas no such verb “to
132
pepsi” was needed to refer to consuming a cola beverage.
Trademark holders may claim that such uses dilute the
distinctiveness of the mark—that to use “xerox” to refer to the act of
photocopying generally, for example, will cause “Xerox” to cease to
function as a trademark. But context and form make all the difference.
Just as the words “Apple” and “apple” can occupy the same lexical space
but denote different referents, the fact that consumers “tivo” a television
program doesn’t necessarily mean that consumers no longer recognize
“TiVo” as a trademark. Webster’s Third New International Dictionary, for
example, defines “linotype” in a way that explicitly recognize polysemic
uses: as meaning “to operate a Linotype machine” or “to set by means of
133
a Linotype machine.” Indeed, the top-of-mind use of a trademark in its
verb form, far from indicating the mark’s generic status, may well
134
indicate the enduring fame of the brand.
This is not to say, of course, that a trademark can never become
135
generic; such a fate has befallen several former marks. But given what
we know about how consumers use marks in conversation, it seems overly
simplistic to guide the genericism inquiry by reference to formalistic
criteria. The reasonable consumer treats trademarks as a part of
136
creating new word forms where those forms enhance
language,
132
1 ALLAN, supra note 27, at 243. Writers on linguistic issues, too, can be in
disagreement about the breadth of such principles. Compare, e.g., SHUY, supra note 17,
at 50–51 (stating that “[o]ne of the signs of the generic nature of an expression is its
ability to flexibly move across grammatical categories”), and Butters, supra note 120,
at 513 n.19 (suggesting that the use of a mark as a verb “would appear to be far more
robust linguistic evidence that the user does not think of the term as a brand name”),
with Ben Yagoda, Parts of Speech, N.Y. TIMES, July 9, 2006, § E (Magazine), at E22 (“Not
only is ‘I googled that hottie’ great publicity for the company, but it’s fresh and funny
and an excellent example of how anthimeria gives English an invigorating slap upside
the head.”). As Ben Zimmer has noted, the first known appearance of the verb
“google” is by one of Google’s cofounders, who used the term “googling” in a listserv
posting in 1998. Ben Zimmer, Googling vs. Bing-ing, VISUAL THESAURUS (Jan. 22, 2010),
http://www.visualthesaurus.com/cm/wordroutes/2147.
133
MORTON, supra note 24, at 222 (discussing entry). A lexicographer might
therefore deem “linotype” to be generic while characterizing “Linotype” as a
trademark.
134
Oded Shenkar & Ephraim Yuchtman-Yaar, Reputation, Image, Prestige, and
Goodwill: An Interdisciplinary Approach to Organizational Standing, 50 HUM. REL. 1361,
1364–65 (1997) (suggesting that the use of “Xerox” and “Federal Express” as verbs
indicates “[t]he highest degree of familiarity” or reputation).
135
Ty Inc. v. Perryman, 306 F.3d 509, 513 (7th Cir. 2002) (giving thermos, yo-yo,
escalator, cellophane, and brassiere as examples).
136
See, e.g., Toyota Motor Sales, U.S.A., Inc. v. Tabari, 610 F.3d 1171, 1180 (9th
Cir. 2010) (“Although trademark holders gain something by pushing their trademark
into the lexicon, they also inevitably lose a measure of control over their mark.”);
Fritinancy: Verb My Noun, http://nancyfriedman.typepad.com/away_with_words/
2006/07/verb_my_noun-1.html (July 14, 2006) (“Companies spend millions turning
their brands into household words—words that take on vigorous independent lives,
stretch their muscles, and through the magic of anthimeria become active parts of
Do Not Delete
2010]
12/15/2010 10:02 PM
THE GRAMMAR OF TRADEMARKS
1349
communication and relying on context and cooperation when necessary
to clarify matters. In many cases, that context will reflect the continuing
vitality of the trademark. Assessing such evidence by whether consumer
discourse complies with certain rules, by contrast, is likely to lead to a
misguided result.
IV. CONCLUSION
Incorporating a greater awareness of how the reasonable person
talks about trademarks does not mean resorting to a linguistic free-for-all.
Trademark owners must, of course, be sensitive to uses by third parties
that indicate that their mark is losing its status as the proper name of a
brand of good or service and instead is becoming generic, such as the use
of the mark by a competitor for its own branded good or service. But
courts, in providing guidance to trademark owners, should be equally
sensitive to language formation and adoption. Greater attention to this
process does not necessarily require additional evidence; rather, it may
simply require a more nuanced and linguistically informed interpretation
of the existing evidence.
Trademark holders may well continue to exhort consumers to talk
about trademarks only in the authorized way—whether through
advertising, entreaties to dictionary editors, or cease-and-desist letters—
but such efforts are unlikely to result in an organic change in consumer
practices. Consumers are likely to resist uses such as “Frosted Flakes
brand cereal” or “Xerox brand photocopier” in typical conversations for
several reasons, not least of which is that such formations violate the
137
In other words, if consumers are
Gricean maxim of Quantity.
communicating satisfactorily with their existing speech patterns, it is
difficult to see why they would be inspired to make things more
complicated, save for fear of liability. Indeed, as linguist Geoffrey
Nunberg has noted, trademarks are part of a universal naming system
that can be remarkably robust even when they are used by speakers
employing dramatically different language variations:
You may not know how to say “soft drink” or “athletic shoe” in
Italian, but nowadays you can always get by in Rome by asking for a
Coca-Cola or Nikes. From an international point of view, those are
the real common nouns now. We’re all drawn together under the
speech—and then spend millions more to keep us from using the words in our
households.”). Cf. Lunsford Jr., supra note 49, at 330 (“‘[T]rademarks are not parts of
speech because they are not properly a part of the language.’” (quoting 1947 letter
from Edward S. Rogers to G. & C. Merriam Co.)).
137
GRICE, supra note 36, at 26–27 (characterizing the maxim of Quantity as
encompassing two submaxims: “Make your contribution as informative as is required
(for the current purposes of the exchange)” and “Do not make your contribution
more informative than is required”). See also LYNCH, supra note 34, at 227 (noting that
speakers of a language don’t stop using a word just because a dictionary indicates that
the use is substandard).
Do Not Delete
1350
12/15/2010 10:02 PM
LEWIS & CLARK LAW REVIEW
[Vol. 14:4
international lingua branda, with only our separate verbs to keep us
138
apart.
Courts would therefore do well to stop encouraging trademark holders to
“police their marks” through such third-party efforts and to focus instead
on brand-recognition advertising and enforcement actions against
139
competitors.
To be sure, adopting a linguistically descriptive view of trademark
validity is only the first step. Recognizing that the use of “xerox” as a verb
does not necessarily mean that “Xerox” has ceased to be a trademark
stops courts from plunging ahead to a finding of generic status, but it
does not tell judges when the tipping point has been reached or whose
usage courts should consider in order to determine when generic status
exists. Those inquiries are, at root, the same as they ever were: Do
consumers understand the word in question to be the proper name of a
brand of product or service or the common name of a product or
service? The answer to this question should continue to be descriptive in
nature but should also be properly focused. If courts are to rely on
presumptions about how the reasonable consumer communicates, they
should at least consider the information that linguists already know in
this regard. The reasonable consumer is not a slave to structure; she can
use language, including trademarks, in diverse and playful ways without
losing sight of her starting point. Modern linguistics theory recognizes
that effective communication can take place even absent prescriptive
rules of grammar. Trademark law, which aims to promote effective
communication in the commercial realm, should take better account of
this insight.
138
GEOFFREY NUNBERG, Branding the Phonetosphere, in THE YEARS OF TALKING
DANGEROUSLY 189, 192 (2009).
139
Folsom & Teply, supra note 28, at 1354 (contending that efforts to police
trademarks are economically inefficient because “such measures normally do not
stimulate demand for a firm’s product”); Higgins, supra note 24, at 389 (“While we
cannot say with certainty that the Coca-Cola Company, for instance, intended merely
to regain its increasingly generic trademark when it marketed its ‘New Coke’ product,
its campaign certainly did have that effect.”).